Introduction

The basic purpose of liberalization is to make the market more competitive and to eliminate the price distortion by making symmetric dissemination of information of the products and services to the market. Thus it will be not possible for any producer to make excessive profit in the market in a systematic manner. An efficient financial institution should offer services to the customer at a competitive price and in a transparent way giving the full information about their services to the customers. Measures to increase competition in the banking sector are an integral part of the process of liberalization of total financial system. The banking sector plays a key role in the economy by functioning as financial intermediaries offering access to payment system, transforming assets and providing portfolio management of various types of investments.  Now lowering the cost of loanable funds encourages investment expenditure in the banking sector and that leads to higher rate of economic growth of a country. Higher efficiency of banking sector naturally, drive out the inefficient banks from the market unless protected by the state and facilitates a transparent and competitive market system. During the course of liberalization it will be important to know the effect of competitive pressures on different types of banks, namely, public sector, private sector and foreign banks. This requires the knowledge of performance of banks in different sectors in terms of associated cost of operation. It has been found that given the scale of operation and product mix the average cost of operation varies significantly from bank to bank. This dispersion in cost basically comes from the inefficient use of inputs and to strengthen the financial sector under liberalized regime it is necessary to understand the forces behind the variation of efficiencies of banks.

Problems of Indian Banks

The financial sector reform in India is a part of the total liberalization process that initiated in the early nineties under the leadership of the then finance minister Dr. Manmohan Singh. It covers the deregulation of financial markets, advances technology, growing customer sophistication, stricter regulation and transfer to undertaking banking activities under flexible interest rate and exchange rate. These reform measures has produced immense effect of the financial markets in India in the form of increasing competition among commercial banks and offering better services and competitive interest rates to the customers. Competition among financial institutions have created higher efficiency in saving mobilization, credit allocation and cost optimization.

Technology is the key issue that changes the banking sector in terms of their efficiency and sustainability. Foreign banks and new private banks are equipped with modern technology and the employees are mostly with compute background with some experience in banking. Whereas, the public sector banks and old private banks are basically depend on old system of operation and the employees have very little knowledge of handling computers. Keeping with the pace of liberalization Government’s stake in some public sector banks is reduced and consequently public equity in those PSB is enlarged. Naturally, management in public sector banks are keen to increase profitability as the public shareholders expect good returns on their equity.

The followings are the recommendation of The Working Group appointed by The Reserve Bank of India, in consultation with the Government of India, on Restructuring Weak Public Sector Banks.

a. Reduction of non-performing assets;
b. Increasing income and reduction of costs.
c. Rationalization of staffing pattern and branch structure.
d. Exploration of the possibility of a limited asset reconstruction fund.
e. Product analysis and product delivery.
f. Information system and business projections.
g. Increased computerization.

Based on the diagnostic studies some of the measures towards restructuring the banks were recommended by the consultants are as follows:
 
a. Total management of NPAs through recovery of NPAs through judiciously negotiated compromise, settlement and invocation of government guarantees.
b. Cost reduction through wage freeze and downsizing through VRS.
c. Improving the business through deposit growth to bring it on par with that of leading banks and better credit delivery through revamping of credit appraisal systems.

Some of the major causes of weakness of public sector banks are as follows.  (i) Operational failures mainly relate to high level and fresh generation of NPAs, (ii) declining market share in key areas of operations, (iii) limited product line and revenue stream, (iv) absence of cost control and effective MIS and costing exercise and (v) overstaffing with low productivity and a high age profile. Restrictive practices in deployment of staff have further aggravated the cost of overstaffing. In many banks, staff cost as a proportion of total operating income is high and the low level of skill and motivation of middle management staff pose a problem in generating business activity to its fuller extent. It has been argued by various committee and commissions set up for evaluating the performance of Indian banks that the weak banks do not have an efficient credit management team and that further advances made by them are most likely to end up as NPAs. Table -1 shows the ratios of Gross NPA to Gross Advance and Net NPA to Net Advance for the years 1997 and 2003 for schedules commercial banks in India. It has been observed that the Gross NPA to Gross advance ratios varies from 7% to high of 39% in 1997 and the range of variation of the ratio remain almost same in the years 2003. However, the Net NPA to Net Advance ratio decreased from the level of 1997 figures and it cradled around 3% in the year 2003. If we compare the figures with advanced countries we find that non-productive loan to total loan in 2004 was around 2%. However in China the figure was around 15.6%, in Malaysia it was 11.6% and in Indonesia it was 13.4% (Mohan, 2005). But to increase the efficiency of financial sector the cost of non-performing asset should be reduced to a comparable figure with the advanced countries.

The reason for creation of NPA in banks is basically the poor quality of credit and that is considered as critical issue relating to macroeconomic structure of the country, competition and banking supervision. It is well known that the financial structure of an economy plays a key role in the allocation of resources and bank credit is a main connection with the real sector. Thus any misallocation of the borrowed fund may cause a huge loss of investment. Moreover, the cost and the availability of bank credit affect the selection of investment both with respect to firms’ financial structure and in relation to the structure of household financial portfolios and banking liabilities.
 
The organizational restructuring should aim at improving governance of the banks and enhancement in management involvement and efficiency. The ratios of staff cost to operating income are found to be high in many public sector banks in the industry. Naturally, these banks have little choice but to take all possible measures to reduce their staff costs and bring it in line with at least the average performing public sector banks in terms of its percentage to total operating income (net interest income + non-interest income). The reduction of staff cost of banks can be done by reducing their expenditure on staff either by reducing the number of employees or by effecting a reduction in per employee cost. The size of the reduction in the number of employees is, however, an issue that has to be addressed very carefully. The reduction will be such that the weak banks could expect to gain the competitive efficiency and a chance to survive as viable units. But at the same time, considering the humanitarian ground, this reduction should be kept at minimum.

It has been argued by different committees set up to investigate the causes of poor performance of public sector banks (PSB) that operational flexibility and functional autonomy of PSBs will definitely improve due to partial privatization. During the post liberalization period Government diluted the holding stake of equity to 51% and it has further proposed to reduce holding to minimum 33% on case-by-case basis. The entry of new private banks and foreign banks will promote competitiveness by introducing new products and better technology.

The committees advocated in favour of deregulation to encourage competition to increase productivity and efficiency. The banks guided by the principle of free market are likely to change product mix, client mix and geographical areas of activities by executing appropriate human resource management given the technological constraint. The banks may opt for more risky asset to earn higher expected return on assets. Banks are likely to shift higher funding cost and interest rate risk to borrowers. The synergic effect of deregulation-induced competition will lead to higher level of efficiency, better resource allocation, innovation of products and progress in technology.

Thus the concerns of Indian banks to cope up with the ongoing financial liberalization, among other things, are primarily two. (i) The large volume of non-performing asset with commercial banks, particularly with the public sector banks and (ii) the problem of excess employees and high staff costs in the domestic banks, particularly, in the public sector banks. Under these circumstances it is necessary to understand the causes of high NPA and the impact of labour congestion on efficiency in Indian banks. In this paper attempts have been made to understand the efficiency of Indian banks of three organizational types, namely, public, private and foreign. Then attempts are made to find out the extent of labour congestion in those banks and to find out the sources of variation of efficiency among the banks during the period 1997 to 2005.

Efficiency of Indian Banks

As a first measure we have used production or value added approach in estimating the cost efficiency of Banks for all the years using Data Envelopment Analysis (DEA). The average figure of efficiencies of the public sector banks shows that the efficiency level is more than 70% for most of the years. The private sector bank show a lower level of efficiency compared to that observed in public sector banks. The values of efficiencies of these banks cradled around 40% during the years of study. The numbers of efficient banks are very low compared to the other two sectors. Foreign banks are much better than the banks in other two types of ownership. The percentage of efficient banks in this ownership type is around 30% of the total number of banks in this sector. The mean efficiency level of all the banks does not show any trend during the period and it lies between .49 and .85 during the period of study. The percentage of efficient banks in total is around 12% in all the years. Thus, the results suggest that public sectors banks are doing better than the private sector banks. However, the levels of efficiency of foreign banks in most of the years are higher compared to the banks in other two sectors.

We have also measure the efficiencies using the intermediation approach and it has been observed that the pattern of efficiencies obtained from these two approaches is similar but the values in the production approach is lower than those of intermediation approach. The efficiency scores in this approach are falling over the period of study across all types of banks and it is more prominent in the case of private sector banks in India. The correlation between the two measures indicates that the efficiency scores measured from these two approaches are highly correlated. The values of correlation of Cost efficiency between these two approaches are higher compared to those of Technical efficiencies.

Input Congestion in Indian Banks

Input congestion means if there is a withdrawal of inputs from the production process then there will be increase of at least one output. It has been observed from the values of input congestion during 1996-97 to 2004-05 that most of the public sector banks have input congestion during this period. There is no sign of reduction in the number of banks with input congestion during this period. However, some of the banks in public sector show no congestion at all over the entire period.

The input congestion is less prominent in the private sector banks. It is also evident that the percentage of private banks with congestion is lower in most of the years compared to the public sector banks. Non-existence of input congestion is discernible for four banks during the period of study. The values show that the percentages of foreign banks with input congestion are less compared to the banks in other two sectors in all the years studied. The Foreign Banks, as expected, show very little or no input congestion during the period.

From the DEA model for sub vector input model and the DEA with strong disposability of all inputs, as defined earlier, we can identify the banks with labour input congestion. The comparison of the percentage of banks with labour congestion among the three groups of banks, namely, Public sector, Private sector and Foreign, reveals that foreign banks have either no congestion or very little labour congestion during the period of study. On the other hand, percentage of Public sector banks with labour congestion varies from 7.4% in the year 2000 to a high of 44.4% in the year 2002. The Chart also indicates that there is a faint increasing trend of the labour congestion in private and public sector banks. The reason may be that during this period most of the banks were computerized but due to pressure from labour organizations and lack of initiative from government it could not be possible to reduce the excess labour forces that to be replaced by introduction of new capital. The VRS scheme once introduced, particularly in the public sector banks did not work well. However, this may be a temporary phenomenon during the adjustment period.

Thus, as expected the labour congestion is high in the public sector banks and the presence of labour congestion is much less in the private sector banks. The foreign banks have very little or no labour congestion during the period of study. Also, the replacement of labour force by introduction of new capital intensive technology are sluggish and there is a trend of higher labour congestion in public and private sector banks, however faint during he period of study.

Factors Influencing Efficiency

It has been observed that there is a wide variation in the efficiency of commercial banks in India. Now to investigate the factors influencing the variation of efficiencies econometric models are estimated with estimated cost efficiency values from DEA approach as dependent variables.

The variables that influence the variation of efficiency can be subdivided into two groups, (a) institutional aspect and (b) bank specific characteristics. The analysis aims at finding out the effect of the ownership pattern of banks, which is categorised as institutional aspect. Three types of ownership are considered, namely, public sector banks, private sector banks and foreign banks. The bank specific characteristics are namely, size, input congestion- both total and labour, and net non-productive asset to advance ratio.

The results suggest that both the cost efficiency and technical efficiency of foreign banks are higher than that of the public sector banks. The result is very natural since the working environment of foreign banks is much better compared to that of public sector banks, and the public sector banks has some limitations in respect to input use and generating outputs. The basic reason of higher efficiency in foreign banks is low labour asset ratios and intensive monitoring of the performance of banks by the authorities, which is possible mainly due to small number of branches of these banks. Contrary to the general belief, the coefficients of the private sector bank dummy show that the efficiencies of private sector banks are significantly lower than that of public sector banks. Since the scale of operation of many private sector banks is lower than that of public sector banks they can not rip out the advantage of scale for being efficient both cost and technical. It has been argued that the public sector banks lend much higher proportion of their loan to the priority sectors compared to private sector banks, and as a result they could be receiving compensating benefits. On the other hand, it is also argued that the conservative behaviour of private banks in lending money in high-risk portfolios make them less efficient because these are basically safer and lower earning ventures. It has been argued by some authors that public sector banks benefited more from the deregulation compared to the private sector banks in terms of cost efficiency. Private Banks tried to transform their mode of operation drastically from labour intensive to capital intensive technology by introducing sophisticated computerizing system and other automatic machines without increasing the effective higher revenue generating activities during this period. Also they have been incurring huge cost in upgrading their employees. Thus to enhance their performance to generate revenue they have performed poor in terms of cost efficiency. The scale of operation, one of the characteristics of banks played a major role in explaining the efficiency of banks. 

The next important bank specific character considered in our model is NPA_AD, i.e., net non-performing loan to total advance of the bank. The significant negative coefficient of this variable suggests that problem loans create additional cost associated with monitoring the utilization of loans and effort to collect the repayment of bad loans from the debtors. Moreover, the cost and the availability of bank credit affects heavily investment choices both with respect to firms financial structure and in relation to the structure of household financial portfolios and banking liabilities. Thus the finding suggests that high level of problem loans usually seen as a signal of the financial distress of a bank and leads to inefficiency of banks. It is implied that quality of loans has some effects on bank’s efficiency.

Labour congestion factor in explaining the variation of cost efficiency is found to be negative. The coefficient of total input congestion is found to be negative and significant for both the models with cost efficiency and technical efficiency. 

Finally, attempts have been made to understand the effect of the level of efficiency and other variables on the level of Non-performing asset of banks. The coefficients of the variables indicate some interesting features of commercial banks. The significantly negative coefficients of private sector bank and foreign bank dummies indicated that the level of NNPA to Advance ratio is much higher in public sector banks compared to the other two groups of banks. This is very natural in the sense that most of the public sector banks do not have an efficient credit management team and that further advances made by them are most likely to end up as NPAs. Also there are gaps in skills, which to be addressed separately on this issue and not by unnecessary curbing growth in advances, which may adversely affect the economy.

It has been fount that here is an inverse relation between size and NPA to advance ratio. It has been mentioned earlier that the scale of operation of public sector banks are higher compared to that of private sector banks. It is also found that the public sector banks use to give a large part of their loans and advances in priority sectors and in most of the cases the repayment of those loans are not done according to the terms and conditions. Thus, the NPA is more prominent in Public sector banks, whose sizes are comparatively high. As a consequence we get an inverse relationship between size and NAP to advance ratio.

It is often argued that high level of NPA is a consequence of inefficient management of banks in disbursing loans and collection of repayments from the borrowers. To verify this argument we have included the cost and technical efficiency in the regression as independent variables. It has been found that both these efficiencies has negative effect on NPA to advance ratio that indicates that lower the efficiency of banks higher will be the NPA. Thus it is evident form the analysis that poor management is one of the causes of high NPA to asset ratio.

Concluding Remarks

The finding of the paper can be summarized as follows. In our study, we have calculated the cost efficiency scores of Indian commercial banks using DEA, and tried to analyse how the efficiency scores across banks of different ownership pattern varies. Average cost efficiency for public sector banks is greater than that of Private sector banks. However, the average efficiencies of foreign banks are much higher compared to the efficiency of banks of other two sectors.

The presence of input congestion in public sector and private sector banks suggests that if a portion of inputs were withdrawn from the process, banks can achieve higher level of output or at least there will be no reduction of output due to the withdrawal of inputs. The public sector banks show high labour congestion during the period and higher than private and foreign banks. This feature supports the notion that public sector banks suffer from over staff. It was expected that after liberalization the banks could use the optimum level of staff to become more efficient and competitive in the market. However, this is not observed even after more than fifteen years of liberalization and this phenomenon persist due to political pressure on the Government and pressure of labour union on the management of banks. It is also argued that since public sector banks have some social obligation they have to absorbed more labour than the banks in other two sectors.