Exclusive: Just how bad the bank NPA problem is

A cursory analysis suggests that at the end of the third quarter, average bad loans for the system will rise to 7 percent from 5.1 percent. That's not all. Most bankers have said the bad loans may see a similar jump in the fourth quarter.

These revelations have led to a scare in the country about how many bank loans will not be paid back and hence, how will the banks survive and lend in the future.

To obtain an answer to these questions, CNBC-TV18's Latha Venkatesh spoke to three experts: Anil Agarwal, Head of Asian Bank Research at Morgan Stanley; Ananda Bhoumik, Chief Analytical Officer at India Ratings; and Dr KC Chakrabarty, Former RBI Deputy Governor and Former Chairman of PNB.

Below is the transcript of the interview on CNBC-TV18.

Q: Ananda, let me begin with you since as the debt expert, you are going to be looking at the bank books from a very long time in terms of how much will come back. Going by what you know about the corporate sector at India Ratings, where does the public sector gross NPA stop at – at 12 percent, at 13 percent? What are you preparing for?

Bhoumik: We have already articulated and everybody knows this, a total stress assets number of about 12 percent is factored in. Where the surprises are coming from is of course, another 5-6 percent in the corporate side, which are currently classified as performing for various reasons and are not provided for at all. And these are assets, loans, exposures in sectors, they are obviously under stress, metals primarily. And our sense is that they will probably need some haircuts.

So, a double digit number is a given. It is really the additional provisioning that people are concerned about and really where that leads to is what sort of internal capital would these bags be able to generate and whether they would have sufficient capital, or any capital at all left for growth. That is the question we are grappling with.

Q: A similar question to you. A blanket question on the public sector bank numbers. We have got a bulk of the public sector bank numbers for this quarter. Is your sense also like Ananda says, that the total gross NPAs will be about 10 percent of the total book, or will it be even worse?

Agarwal: We look at total impaired loans, not just gross NPAs. So, if I look at a quarter back, the total impaired loans in the state owned banks was about 14 percent. There are banks which are running at almost 20 percent. That 14 percent will probably go up by at least 300-400 basis points over the next two years.

We also did a stress test recently, where we basically built in a scenario where it goes up by about 800 basis points compared to where we are right now. So, that 14 percent is just definitely going to go up materially whether it goes up to 16-17 or 21-22 percent, we need to wait and see how the economy recovers.

Q: Going by the pace of recognition and the kind of recognition that banks have done in the just concluded results season, what do you think will be the recognised gross NPA by the end of March?

Agarwal: Recognised gross NPAs, if you look at RBI's asset quality review (AQR) for most of the banks, it is about 2-2.2 percent. If you add up 1 percent, the gross NPA ratio will probably, for the state owned banks as a whole, increase by around 2-2.5 percent between December and March quarters.

Q: I am trying to get that figure of the gross NPA, because ultimately provisioning will have to be done on stated NPAs, not on expected NPAs. So, if that is case, what kind of a capital requirement do you envisage for the state owned banks for FY17?

Agarwal: The question is, what we do is that we look at NPAs, we look at restructured loans, because if you look at the floor of restructured loans right now, for the last couple of years, a large chunk of the restructured loans are turning bad. And what we do is we look at the restructured loans, which are coming out of restructuring and 60 percent of those loans actually becoming NPAs...

Q: Actually, it is only between about 30 percent at worst or 40 percent at worst of the restructured loans that have become NPAs. So, should you count the entire restructured book as potential NPAs?

Agarwal: No, we do not count the entire restructured book as potential NPAs. Actually, if we look at the loans which are exiting the restructuring process, most of the banks right now are reporting 60-80 percent of their loans which are exiting restructuring process, becoming NPAs because if you look at 20-30 percent on the outstanding stock, that is not the right number because a lot of these restructured loans have moratorium periods of 2-3 years. So, you need to look at what is coming out of restructured loans. 

When it comes out of restructured loans, is it getting upgraded or downgraded. And depending on the bank, this quarter is different because of the special drawing rights (SDR) stuff, because most of the restructured loans are being, whichever, whatever is exiting is basically going through the SDR process, so they will become NPAs two years later.

But, if you look at the exit from restructuring and that is the key number in my view, 60-80 percent of all the exits from restructuring are turning bad. And that applies to private and state owned banks. So, a lot of restructured loans are turning bad. As a result of which, when the time comes, they will have to make provisioning on those restructured loans also.

So, in my view, the state owned banks, if I assume that they need to achieve 9.5-10 percent comment equity tier I (CET1) ratio, they need to grow their loan book at anything between 7-10 percent depending on how small or big the bank is. Then to achieve that kind of Tier I ratio by March, 2019 which is the end of Basel  III, state owned banks will need about USD 40 billion of capital. And, the thing is the more the delay in capital infusion, the bigger the hit will be because, that means these banks are not able to grow, so more corporates will turn bad. So, it is not a static number, it will keep growing.

Q: That is a sizeable amount of money that you are asking for in terms of capital. In rupee terms, it will be about Rs 2.5 lakh crore, that you are talking of in terms of capital. What is your bill? And I want to know more the FY17 bill.

Bhoumik: FY17 is going to be relatively modest. You see, the complication here is you were discussing provisions is not so much as what the addition to the reported NPAs are. The other complication has been that this year and rightly so, some of these recognitions are being provided for as if in they are in the doubtful category. So, RBI has required them to be classified from the day they went under stress and not from recognition. So, therefore, some of these recent recognitions are requiring higher levels of provision which is what they should be. The FY17 requirements are relatively more modest. But, FY18 and FY19 are significantly stepped up. At FY 17, depending upon which way you look at it, the growth numbers you look at, we sense a requirement between Rs 25,000 crore and Rs 30,000 crore. That sounds modest. Q: Rs 25,000-30,000 crore?

Bhoumik: Of equity contribution from the government, so I am just looking at what the government is required to contribute. On top of that, there is an additional 25,000 that the banks need to raise by way of ET1 capital. So, somebody has got to find a market for that. If that is not supplied in, then obviously, that would probably need to be contributed by way of equity and then of course, there is a balance 20-25 percent number which is expected to be raised to from the equity markets which is not supportive at all.

So, therefore, in terms of the actually billing, that number may actually double from what the ask is. And this is without even considering the additional provisioning that may be required from the bottom up study that we did which suggests that there is an additional haircut that the banks need to take on some of the currently performing numbers. So, the bill is going to mount up._PAGEBREAK_

Q: Let me look at what the equity market is expecting. If the Budget were to indeed say Rs 25,000 crore or Rs 30,000 crore, how inadequate would that be? What is the number for FY17 that you want the Budget to say in terms of recapitalisation?

Agarwal: Again, we do not have a Budget number per se, that we have published so I cannot talk about that per se, but as I said, following on Ananda’s point, because he is breaking it down to three years. My view is that the total bill is a big number. If I look at any of the banking slowdowns or banking sectors which have gone through an NPA problem in the past, the out for the banks or the bank stocks really started performing only when the investors were sure that the balance sheet is taken care of. The share count is fine.

So, if you look at US banks in 2008-2009 crisis, there was a massive capital infusion, one shot. There were a number of capital infusions through 2007 and 2008, but towards the end of 2008, there was a massive capital infusion. March 2009, investors were like, this is a share count. They do not need dilute anymore and the stocks went to the races. 

For the banks also to perform, market has to see how these banks will be able to raise a significant chunk of this USD 40 billion as quickly as possible.

Q: If the government could give a lumpsum, probably a Rs 50,000 crore or Rs 60,000 crore in FY17 itself, if a big number is announced, do you think that that will make life a little easier to raise money from the equity markets? Would an upfront big payment work?

Agarwal: It should help by a bit. I do not think it will remove the problems completely, but it will help to a small extent.

Q: We have just discovered that at the end of Q3 most of the banks, the better ones, are reporting about 5-7 percent in terms of gross NPLs and the weaker ones are reporting between 8-10 percent as gross NPLs. In your own estimate how bad can the picture get? By the end of Q4 do you think we will get an average for the public sector banks of about 10 percent GNPAs?

Chakrabarty: If we take into account the amount, which has been technically written off in the last five to six years, already this amount which I consider technical write off is as good as gross NPA because banks make equal effort to recover this amount.

Already GNPAs of many banks is already more than 10 percent. I don’t know what will happen to the restructure but safely we can say that stress assets of the banking system will be around 20 percent, GNPA should be around 13-14 percent but then NPA is not lost. 

We must understand, how much will be the NPA that is not important, it may be 16 percent, it may be 20 percent, what is important is what banks are doing to recover this NPA and what banks are doing to ensure that further slippages are not taking place. That is more important, this number is not that important.

Even if it is 20 percent, banking system will be able to absorb the shock. However, if they don’t stop the process of creating the NPA and they don’t know how to resolve the existing 20 percent of the GNPAs then we have a problem. 

Q: At the moment the government has said that it will give above Rs 25,000 crore in terms of capital in FY17. Do you think it would be wiser for the government to probably double that amount simply because that will mean more confidence for the banks to raise equity from the capital markets or from the bond markets, what do you think should be the best strategy from the shareholder, that is the government? 

Chakrabarty: Rs 25,000 crore is peanuts. We must understand that a bank requires capital not only for the provision it is making for the NPA; banks require capital to expand new credit. Banks also need to meet the Basel III norms. Now, if we consider all these three things together, banks capital requirement is much higher. 

It will be wiser for the government to privatise some part of these banks after taking the knock of these NPAs and whatever is the market determined price at that price the new equity should be raised.

However, before that you have to make thorough reform of the administrative structure of the public sector bank, their recruitment pattern, their management pattern and also bring some sense of accountability. By giving ad hoc Rs 25,000 crore, it is not going to solve the problem. Even Rs 50,000 crore is not going to solve the problem.

Q: What you say is definitely the ideal picture but that would require legislative reforms, legislative changes especially if the equity stake has to be brought down. Given the realities of the day in terms of the way Parliament is, what do you think is the best course that the government can adopt for FY17, both in terms of capital and in terms of reform measures?

Chakrabarty: Up to 51 percent, the entire equity government can give but then it will affect the quality of growth of the banks. Please understand, if we have to get out of this situation, banks have to do new business. Now, if we are saying that because of the legislative problems, okay, then banks have to go and raise the equity in the market; remaining 50 percent. I see for the last five years and the government is contributing to the equity capital, there is no private raising of the capital.  We have to understand that and from where government is giving the money? Government has a fiscal deficit. By borrowing the money from the public and giving the money to the public sector banks, I don’t think this is a wise step. If government has the fiscal surplus I have no problem but they are borrowing and borrowing to invest in the equity, I think that is the worse type of financial brinkmanship. 

Q: That is absolutely correct but the problem still is that there is a legislative deadlock when it comes to being able to rewrite laws or amend laws. Given that reality, that the Banking Regulation Act cannot be amended in a jiffy what may be the best course available? Do you think that the government should provide some kind of a bond related money to the public sector banks and then perhaps open doors for equity? 

Chakrabarty: Whatever I say, what I am saying banks have to raise fresh equity capital; absolutely there is nothing. Bonds, yes, they can salvage to some extent but what we are saying is if banks do not get equity, growth will suffer. Then you have to be ready, then you have to give new banks licences so that this business would be shift to non-public sector banks. 

We should be very clear, if the public sector banks are going to continue, public sector banking will be not there. Let us be very clear what is the present situation, you cannot willy-nilly and say that legislative problem, this problem, that problem. I don’t think if there is a problem of regulation then we must amend the legislation. This is an urgency situation, if it does not happen then public sector banks growth will suffer. 

Q: What is the best financing route available to the government under the political realities of the day? Is there some kind of bond financing that they can do and later convert that into shares?

Bhoumik: That certainly is an option. The other option of course is which I think has been tried earlier is to work out a long-term financing arrangement with some of the multilateral agencies like World Bank and relate that to reforms in the PSU bank which I think the government is committed to. So, the government may either take directly or may encourage the World Bank to pump in quasi equity or long-term bonds into these PSU banks. 

However, in essence it would probably call for some sort of fiscal expansion because the current amount that has been allocated for the Indradhanush may not be adequate to serve the purpose as KC Chakrabarty said both providing and for growth capital. So, it is probably good to already start articulating a long-term plan, upfront some of the money as you said and then promise more if required in FY18-FY19.