Sunday, July 05, 2015

Balance sheet problems of the firms and the banks

While the official GDP numbers are showing an optimistic picture, trusted firm databases are suggesting that the recession which began in Q1 2012 has not ended.

An important dimension of this recession is the leverage of firms. On an international scale, there is increased interest in the debt `super cycle' which appears to play out distinctly from the business cycle. There are periods where households and firms are adding leverage, and these tend to be a good time for the economy. And there are periods where balance sheets are stretched, and this yields a drag on growth.

The credit boom from 9/2003 to 9/2008

This viewpoint emphasises the importance of credit booms. The great Indian credit boom took place in Y. V. Reddy's period as governor, where the pursuit of exchange rate policy gave lax monetary policy:

Figure 1: Nominal year-on-year growth of non-food credit of the banking system

The dashed lines focus on this period. Year-on-year credit growth peaked at near 40 per cent. This was the biggest-ever credit boom in India's history.

Such periods suffer from two problems. First, there is a surge in the quantity of loans being given out by banks. In such times, the level of scrutiny tends to go down. When the surge happens at the time of a business cycle expansion, it's particularly easy to be over-optimistic. Such a drastic credit surge was going to result in trouble.

These events are quite some distance away in time. This is a different time scale when compared with business cycle fluctuations. However, it appears that this credit surge matters to understanding what holds the Indian economy back today.

Credit distress of the firms

From 9/2003 to 9/2008, a lot of debt was taken on by firms. It is likely that some of this capital was misallocated by banks who gave loans to unworthy firms. Soon after that came the Lehman crisis, and soon after that the current recession began (in Q1 2012). Recessionary conditions have hampered profitability of many firms. For many firms, the combination of low profits and high debt has generated credit stress.

Figure 2: Interest cover ratio (ICR) of all non-financial firms observed in the CMIE database

The graph above shows the interest cover ratio (ICR), defined as PBDIT/interest. A firm at an interest cover ratio of 1 is in a lot of trouble: All it's operating profit is taken in paying interest.In the graph above, the blue line uses annual data, where the accounting data is of higher quality, and the black line uses quarterly data, where the data lag is lower.

At its peak, the average interest cover ratio was at 10, which means that the operating profit was 10 times the interest payment. There has been a dramatic decline in this ratio, with increased interest payments and reduced operating profit. Unlisted companies have consistently been under significant credit stress through this entire period; this low value of the interest cover ratio and its lack of time-series variation is a mystery that merits further exploration.

Figure 3: The fraction of total assets in firms where ICR < 1

Most people will agree that by the time the interest cover ratio has hit 1, the firm is in significant credit distress. The graph above shows the share of these distressed firms in the overall balance sheet size of India's large firms. This shows a period of acute distress in the late 1990s and early 2000s, where over a third of the corporate sector was in difficulty. Things got dramatically better by 2008; only 15% of the corporate sector was in credit distress. From that bottom, there's been a doubling. For the year ended 2013-14, 29.16% of the balance sheet of the corporate sector is in firms where the interest cover ratio is worse than 1. Roughly 20% of bank lending is stuck in these firms.

In a few months, we will know the situation for 2014-15. However, Figure 2 suggests that things have deteriorated when compared with 2013-14.

Credit distress and fixed investment

In an ideal world, capital should flow to firms with good projects, regardless of their present financial condition. This is not how the Indian financial system works. When a firm gets into credit difficulties, it appears to lose access to external capital. Low profits obviously hamper internal capital.

We pool all CMIE data from 1989 to 2015 and construct quartiles by the interest cover ratio, and within each quartile, report the sample mean of the growth in fixed assets (in real terms):

ICR quartileReal GFA growth
Low (ICR < 0.8)7%
Q2 (ICR from 0.7 to 1.7)10.4%
Q3 (ICR from 1.7 to 3.17)14.5%
High (ICR > 3.17)16.6%

Gross fixed assets (GFA) growth has been 16.6% in real terms for firms where the ICR was above 3.17. For the firms where the ICR was below 0.8, it was 7%. This suggests that credit distress as measured by the ICR is correlated with reduced investment.

By this reasoning, when a third of the balance sheet size of Indian firms is at an ICR < 1, this would result in a drag in investment.

Where are your borrower's yachts?

Unhealthy borrowers give unhealthy banks. Banks in India have roughly 12x leverage. When loans get into trouble, the recovery rate (correctly calculated) is probably around 25%. This means that if a bank has NPAs of 10% of total assets, then there is substantial stress. It will lose 7.5% of total assets, which will be roughly as big as the equity capital.

Figure 4: The state of bank capital

The graph above focuses on the data being reported by banks. RBI and the banks are using various methods to hide bad news, including restructuring and CDR. At the bottom is official NPA data, which is below 2% of total assets. But when we add in CDR and `restructured' assets, this comes up to 6% of total assets. Equity capital is near 8% of total assets.

These are averages for the banking system as a whole. There are undoubtedly some banks who are better than the average, and there are some banks who are worse off.

The evidence above emphasises net NPAs. In this speech on 5 May, Mr. Mundra offers facts about gross NPAs which are of course larger.

The phenomena reported on in this blog post -- firms with credit distress, that are likely to have impaired investment activity, and are likely to face difficulties in repaying to banks -- are likely to be related to the phenomenon of `stalled projects' as seen in the CMIE Capex database.

The true extent of difficulties at banks may be larger than 6% of total assets, as RBI and the banks are collaborating in hiding bad news.

Here is one anecdote. Haldia Petrochemicals is a large firm. They owe banks money and have not repaid. However, this is not classified as a non-performing asset owing to an instruction from RBI. As the story by Pranav Nambiar in the Financial Express of 26 May 2015 says:

Ashutosh Bose, CFO & executive VP, HPL, confirmed to FE the central bank has provided the special dispensation enabling bankers to treat the HPL exposure as a standard account. “We will not be treated as an NPA and banks will not have to make any provisions on our account. The special dispensation by the RBI has been given due to the nature of our business which is impacted by extraneous factors like changes in naptha prices,”

We have a banking regulator who thinks that changes in naptha prices somehow change the fact that this is a non-performing asset. We have a banking regulator who has the power to instruct banks on a transaction-by-transaction basis, under complete opacity, to disregard the (subordinate) law. We have a banking regulator who is willing to use such powers.

This is just one anecdote, where the news spilled out into the open. There is no data, in the public domain, about what "special dispensations" have been given out by RBI.


The credit boom of 2003-2008 and its aftermath are an important element of understanding India's macroeconomic predicament. The balance sheet difficulties of banks and their borrowers are an important part of what happens from here on. Credit distress in India today is not as bad as it was in the late 1990s. But for perhaps a third of the corporate balance sheet, there is significant credit stress.

The key thing to watch is the growth of operating profits. If PBDIT grows, then the ICR will come back into shape. From Q1 2012 onwards, we have been in a recession, and operating profit growth has stalled. Many firms are in a debt spiral where interest payments go up, operating profit fares poorly, new loans are taken to keep the ship aloft, interest payments go further up, and so on.

There may be a bit of a feedback loop going. The economy is bad, so borrowers are faring badly, so the banks are facing difficulties (even with a sympathetic regulator), so the banks give less credit to healthy firms, which further hampers the recovery. This raises the possibility of `Japanisation' where bad news is hidden, firms and banks are absorbed in dealing with credit distress, many long years go by with sluggish growth.

We must learn more about Japan, and other unhappy episodes where the debt overhang hampered the revival of growth in a sustained way. We must also learn about how India bounced back from the last credit boom of the mid 1990s. As the graphs above show, the downturn after this boom in the late 1990s featured credit distress that was worse than what we see today. There was a lot of creative destruction. Large numbers of non-financial firms went under. Their exit, and the operational improvements of the survivors, set the stage for the revival of the economy.

Policy makers need to nudge things away from the Japanisation scenario. This requires working on four fronts:

  • How to create new channels through which capital goes into healthy firms without the involvement of banks,
  • How to be less like Japan or China, where bad news is hidden and incumbent banks and firms enjoy business as usual, 
  • How to create capabilities at RBI for technically sound regulation and supervision,
  • How to resolve failed firms and banks, so as to shift capital and labour to healthy organisations.

Every now and then, there is a call for a bailout of banks and their borrowers. We need to be careful in how we approach this. The scale of the problem is large, and the available fiscal space is limited. Do we want to throw good money after bad? If regulation and supervision has failed before, why will it not fail again?

1 comment:

  1. Well written article.
    The Npa plus restructured assets including cdr in my view is much higher around 15% or so.
    Time to bite the bullet rather than postpone the bad news?


Please note: Comments are moderated. Only civilised conversation is permitted on this blog. Criticising me is perfectly okay; uncivilised language is not. I delete any comment which is spam, has personal attacks against anyone, or uses foul language. I delete any comment which does not contribute to the intellectual discussion about the blog article in question.

Please note: LaTeX mathematics works. This means that if you want to say $10 you have to say \$10.