SUNANDO ROY – On Banking, Finance and Society

Sunando Roy 2006

Introduction

High volatility in asset prices is worrisome because, when financial unbalances unwind, the real economy is exposed to a substantial economic downturn and, very frequently, to recession and possible deflation.

Most industrial countries experienced sometimes violent boom and bust cycles in credit markets in the late 1980s and early 1990s (Hofmann, 2001). Similarly, Latin American countries experienced a boom and bust cycle in both, property prices and credit aggregates. There is clear evidence that in Latin American countries during the 1990s, asset prices booms contributed to an increased perception among individuals about positives changes in wealth and their creditworthiness that allowed agents to increase their indebtedness (Herrera and Perry, 2003).

Likewise, housing prices are more likely to end in a bust and to be costly. For example, Helbling and Terrones (2003) find that only 25 per cent of the equity-price booms in the past 30 years ended in busts, while around 50 per cent of house price booms ended in bust. Policy-makers should also be aware of the fact that bubbles in house prices are more worrisome that those in equity prices, in part, because these tend to reflect domestic credit conditions, whereas equity prices tend to reflect global forces (Selody and Wilkins, 2004).

Housing price bubbles give home buyers a false sense of the real return they can expect on their investment, which can lead to speculative home buying and overinvestment in the real housing stock. This can lead to overinvestment in physical capital, overconsumption, and overextension of credit. The process reverses when the bubble burst. The decline in housing prices results in a deterioration in balance sheets that constrains spending and investment. Given a lower value of collateral, defaults start and  the financial institutions are less willing to lend. This can cause decreased investment and increased bankruptcies.

II

Real Estate Exposures of Commercial Banks in India :

A System Level Assessment

 In the Indian Context, for the past two years we are witnessing an acceleration of credit to commercial sector. Non-food credit, on a year-on-year basis, registered a growth of 30.8 per cent as on March 31, 2006 on top of a base as high as 28.8 per cent a year ago. The expansion of bank credit to  housing sector was strong, benefiting from low interest rates and tax incentives. Credit to the commercial real estate increased sharply, although it still constitutes only a small part – less than two per cent of outstanding non food bank credit . The non-performing component of housing loans by commercial banks at 2.1 per cent as on December 2005 is still at a moderate level. ( Table 1)

Table 1:Growth in Retail credit and commercial real estate

( Rs. Crore)

Source : OSMOS .

Particulars	Outstanding as at the end (All banks			Non performing loans (%)
	Mar-04	Mar-05	Dec-05	Mar-04	Mar-05	Dec-05
Housing loan	89449	134276	168213	1.9	1.9	2.1
Consumer durable	6256	3810	3763	6.6	9.6	9.0
Credit Card	6167	8405	11534	6.3	7.9	7.3
Car Loan	NA	35043	54552	NA	2.0	1.9
Other Personal loans	87170	85077	109239	2.6	3.8	3.4
Gross Retail loans	189042	266610	347300	2.5	2.8	2.7
Growth in gross retail loans	NA	41.0	30.3
Total retail loans as % of Gross credit	22.0	23.7	26.0
Total Gross credit	859092	1125056	1334940
Memo:
Commercial Real estate loans		16419	37877
Total real estate including housing loans		150695	206090

As the total real estate exposure  including Housing Loans jump more than 33 per cent since March 2005 and thereby crossing            Rs. 2,06,090 crore, there is a need to take a look at the emerging risks at the real estate sector. The following analysis is an exploration into the risks present in the Indian real estate market.

Data Analysis

For the purpose of analysis, quarterly OSMOS data have been used.  All banks real estate exposures for December 2005 has been normalized by their asset size.  

Measures of  Central Tendency

The risk in this market seems to be rather low when looked in terms of averages. The mean exposure of commercial banks is highest in housing loans at 3.69 per cent of assets. The rest of the exposures in commercial real estate, mortgages and land and building development add up to another 1.55 per cent. However, it is also apparent that the maximum exposure by a single bank in the real estate market is much higher.

Table 2 :  Scheduled Commercial Banks’ Average Exposure in Real Estate Market

( as a proportion of assets)

	Commercial Real estate	Individual Housing Loans	Land and Buiding	Mortgages
Mean	0.78	3.69	0.55	0.20

Note : Disaggregated data on Banks have been used  by  calculating the real estate exposures in proportion to their total assets

Measures of Dispersion

Measures of dispersion tell us how the risks are distributed. It tells us whether such risks are concentrated in a few banks, which is a cause of concern for policy authorities or they are well diversified.  Measures of variability such as the Standard distribution show greater variability in bank exposure to real estate sector.

Higher kurtosis points towards thicker tails in the data implying more extreme values in the data than that can be accurately predicted by the normality assumptions.

Table 3 : Measures of Variability and Extreme Values

	Commercial Real estate	Individual Housing Loans	Land and Building	Mortgages
Skewness	3.51	1.52	1.61	3.76
Kurtosis	17.16	6.60	4.88	19.59
Standard Deviation	1.59	3.89	0.78	0.47

Correlation Structure

There is hardly any significant association between the commercial real estate loans given by banks and the individual housing loans given by Banks. If the same set of Banks were providing both types of loans, there would be greater risk faced by those banks in the event of an asset price shock. However, for the present, the risks seem to be well distributed.  Significantly, commercial real estate loans have very little relation with housing loans ( a correlation coefficient of -0.01), indicating different sets of banks financing the two fast growing  activities.

Table 4: Correlation among types of Real Estate Exposures

	Commercial Real Estate	Individual Housing Loans	Land and Building	Mortgages
Commercial Real Estate	1.00	-0.01	0.17	-0.06
Individual Housing Loans	-0.01	1.00	0.28	0.20
Land and Building	0.17	0.28	1.00	0.09
Mortgages	-0.06	0.20	0.09	1.00

Distributional Properties in Bank Exposure to Real Estate

Theoretical quantiles are effective tests of divergence from normality of distribution. The absence of normality evident in the Quantile- Quantile graphs below show us that there are more banks at the tail of the distribution than the normal distribution would permit.  With a view to identify the Banks that face greater risk in the real estate sector as compared to the other banks, the superevisor would need to monitor the more aggressive  banks in the real estate market  more closely. This can be critical input to the system of surveillance of Banks.

III.

Performance Analysis of Individual Banks

In real estate market  , there are at present four  categories of Banks, Overaggressive ( exposed by more than 20 per cent of asset size), moderately aggressive ( exposure in Real estate including housing loans between 10-15 per cent of assets ), cautious  ( 4-10 per cent) and dormant( less than 2 per cent).

A,. Overagressive

More than 20 per ent of assets deployed in real estate and individual housing taken together. Two Banks fall in this category.

!   ICICI BANK ( 24.72 per cent of asset in real estate market)

2.. Standard Chartered Bank ( 21.1 per cent of assets in the real estate market)

Clearly, these two bank’s exposure in the real estate market require further exploration. Concerned Bank Monitoring Divisions may be advised suitably in this regard.

B. Moderately Aggressive ,

Exposure to real estate ( including individual housing) at 10-15 per cent of their assets. They should be monitored as they may become very aggressive in near future,  The following six banks fall in this category:

Capital Local Area Bank Ltd	14.72
Hongkong & Shanghai Banking Corporation	13.89
Vijaya Bank	13.27
ABN AMRO Bank N.V.	12.08
Corporation Bank	11.59
Citibank N.A.	10.23

 It is proposed that RBI monitors on a quarterly basis  whether  the exposures  are increasing in the market.

3. Cautious Banks

Majority of Banks( 46) have exposures in the real estate market to the tune of  4-10 per cent of their assets. These are cautious banks. It is proposed that OSMOS monitors on a quarterly basis  whether some of them are migrating to the moderately aggressive category.

4. Dormants

 Rest of the Banks ( 36 in number) are exposed to the extent of  less than 2 per  cent of assets in real estate.  No action is required.  The graph below shows the distribution of real estate exposure ( including housing)

————————————————————————————————————