India's schizophrenic banks

In a move that is commendable, the Reserve Bank of India (RBI) has decided to continue with its recent practice of issuing periodic Financial Stability Reports (FSRs), or assessments of the strength and resilience of the financial system. Last year, reports were issued in March and December. Starting this year, biannual reports are to be issued in June and December.

The June 2011 report reveals much that is known about the Indian financial system: that it is still dominated by banking, that banks rely largely on deposits for their funds, that the allocation of funds pointed to stability, and that the banks were on average well capitalised.

Deposits accounted for 79 per cent of total liabilities while advances and investments constituted 87 per cent of total assets, with investments alone amounting to 30 per cent. Since government securities are an important component of investments, they substantially shored up the balance sheets of banks.

Despite these features of the banking system, the RBI's report is characterised by a muted sense of concern. The reasons for that concern are specified, though often the exact numbers involved are difficult to glean because they appear in unlabelled graphs.

Credit boom

The first of the RBI's causes for concern is the evidence that in recent times banks seem willing to accommodate borrowers, even if that required them to rely on more costly funds mobilised by issuing certificates of deposit or through borrowing. The share of CDs and borrowing in the total liabilities of banks rose from around 7 per cent in the middle of 2009 to 10 per cent at the end of March 2011. This reliance on higher cost funds was the result of an increased proclivity to lend, resulting in periodic credit booms.

Credit growth rose sharply to 22.6 per cent in 2010-11, which called for caution since past experience shows that the process of impairment of assets begins during a credit boom. Further, besides the fact that these funds were costlier than conventional deposits, they were often characterised by short maturities, leading to increasing maturity mismatches between the sources and uses of funds.

To quote the FSR: “While more deposits than advances were getting re-priced in the near-term (less than a year) bucket, more advances than deposits were maturing in 1-3 year and 3-5 years buckets.” This kind of a credit boom, experience from elsewhere suggests, can result in an accumulation of excessive risk and an increase in bank fragility.

What is reassuring is that while this was the tendency at the margin, these kinds of funds were a small share of the stock of resources with the banks, with low cost current and savings deposits accounting for 35 per cent of total deposits.

The causes for concern were not restricted to the pace of expansion of credit and the pattern of fund raising by banks. They also came from the sectoral composition of credit expansion, with incremental credit concentrated on a few sectors, especially sensitive ones such as retail lending (including housing), commercial real estate and infrastructure. Here too there was a difference between the stock of credit assets created by banks and changes at the margin. While on an average, the credit portfolio of banks was diversified across sectors and geographies, in recent years the sectors named earlier have gained in prominence. The combined credit to these sectors increased by 27.5 per cent in 2010-11, as compared with the aggregate credit expansion of 22.6 per cent (Chart 1).

Their combined contribution to the increment in gross outstanding credit between the end of March 2009 and the end of March 2011 was 40 per cent. In the event, at the end of March 2011, the retail, commercial real estate and infrastructure sectors accounted for 19, 4 and 13 per cent of the gross advances of the scheduled commercial banks. As Chart 2 shows, residential mortgages and infrastructure were especially important targets of SCB lending.

Realty lending

On the surface, it appears that lending to the real estate sector should not give much cause for concern. The share of non-performing assets in the real estate sector relative to total NPAs was at 15 per cent, lower than the sector's share in total advances of 17.7 per cent. But things seem to be changing. The rate of growth of NPAs in the real estate sector was at 19.8 per cent, significantly higher than the rate of growth of aggregate NPAs of 14.8 per cent. Moreover, NPAs in the commercial real estate segment grew at an astounding 70.3 per cent. With many banks, including public sector banks having attracted borrowers with schemes such as those involving low teaser interest rates in the initial years, and interest rates on the rise in the economy, this does increase the risk of loan impairment in the sector. Besides residential mortgage, risks abound in the remaining part of the retail lending segment as well, though that accounts for a small share of total lending. Those loans are significantly riskier and largely unsecured. Yet, such lending has been on the rise, given the higher interest rates that can be charged for them.

Surprising trends

Finally, the surprising trend is with respect to bank exposure to the infrastructural sector. Lending here is largely to the power, roads and ports and telecommunications sectors. These are areas where, post-liberalisation, private entry has been sudden and substantial, resulting in huge demand for credit. Banks, including private banks have chosen to step in, resulting in the sector accounting for a significant share of SCB advances. Power alone accounted for 42 per cent of aggregate infrastructure credit at the end of March 2011, with the other two sectors garnering 18 per cent each (Chart 3).

This sets up two kinds of risk. First, the excessive exposure of banks to a few of these sectors, when the aggregate level of exposure is by no means small, is a source of enhanced risk.

Second, given the long gestation lags associated with these projects, which can be worsened by delays in project implementation, commercial bank lending to them is bound to be associated with significant asset-liability mismatches and the associated risks.

There are three features of bank lending to infrastructure that need to be noted. First, as of now the ratio of NPAs to advances in this sector is low, amounting to 0.5 per cent. But that is partly because significant bank lending to this sector is recent. It is likely that the contribution of this area to NPAs would increase over time. Thus, in 2010-11, there was an increase of 42.5 per cent in the impaired loans to the infrastructure sector. Second, it is true that many projects have a guarantee of returns to investment. But this is true mainly in the power sector and that too for the fast track power projects.

Concentrated among PSBs

Finally, exposure to the infrastructure structure is concentrated among public sector banks, which account for 84.8 per cent of banking sector exposure to these industries. Hence, it may be attributed to government policy rather than autonomous bank behaviour.

This, however, does not reduce the risk of default and of resulting fragility and failure, especially since lending is directed significantly at private sector firms. Moreover, in recent times the exposure of the new private banks and foreign banks to this sector has risen significantly, indicating that the “dynamism” in this newly liberalised sector is also an explanation for bank interest.

Thus, as of today bank behaviour in India appears almost schizophrenic, with the evidence pointing to both caution and an increased appetite for risk. In the aggregate, banks appear cautious and restrained with a funding base and asset portfolio that point to resilience and capital adequacy ratios that are more than adequate.

But at the margin, they display behavioural characteristics that point to increased risk-taking of a kind that could lead to fragility and failure, resulting in the regulator's concern, however incipient. Clearly, caution is a legacy, while risky behaviour is the new norm. Concern is, therefore, warranted.

The possible reasons

What could explain this behaviour? One obvious explanation is the quest for profit that encourages players, public and private and big and small, to diversify in favour of sectors such as retail lending and real estate. But that alone cannot explain the change in this direction.

The change has clearly been influenced by liberalisation that allows banks, public and private, to behave in this manner. So long as capital adequacy ratios are within prescribed ranges, regulation does not prevent or significantly constrain behaviour of this kind. A third explanation is the inadequacy of opportunities to lend at a profit to the commodity-producing sectors that are languishing.

The consequence is increased lending to units in the services sector and to infrastructure, besides the retail segment. Finally, there is the demonstration effect. With public sector banks still dominating the banking space, it may be expected that legacy behaviour would dominate the new tendencies.

But the example set by the new private sector banks and foreign banks in residential mortgage and retail lending and in lending to commercial real estate, not just encourages but in fact forces public sector banks to do the same. In the event, in certain areas, such as the provision of teaser loans, the public sector banks are willing to go even further.

The observed outcome is a result of all of these factors and more.

COMMENT NOW