Non-parametric Analysis of Technical Efficiency of Public Sector Banks (PSBs) in India

Abstract

The aim of the article is to analyze and evaluate technical efficiency scores of public sector banks (PSBs) in India. The study also determines the nature of return to scale (RTS) of individual banks and thereby identifies the leaders and laggards in the PSBs. To measure the extent of overall technical efficiency, pure technical efficiency and scale efficiency (SE), the non-parametric approach, that is, data envelopment analysis (DEA) is used to determine the causes of inefficiency. The sample of the study includes 26 PSBs operating in India during the time period from 2007–2008 to 2011–2012. The results show that although the PSBs have more or less similar efficiency scores, that is, higher than 0.900, still out of 22 banks falling in the category of efficient banks in 2007–2008, only 7 of them were left by the year 2011–2012. Overall analysis of PSBs during the time period of the study explains that a greater part of inefficiency among PSBs is attributed to scale inefficiency. In addition, the number of banks operating at constant return to scale (CRS) came down to 9 in 2011–2012 from 23 in 2007–2008. Simultaneously, there was a reduction in leaders and increase in laggards. It is suggested that banks must optimize their scale of operations and adopt technological innovations.

Keywords

Non-parametric approach data envelopment analysis technical efficiency public sector banks (PSBs)

Introduction

Today, after 22 years of liberalization, India has an exhilarating banking sector. Liberalization and deregulation process have made a sea change in the banking system. From a fully regulated environment, banking has gradually moved into a market-driven competitive system. Increasing competition is fragmenting the share of profitability of banks and forcing them to work efficiently. Public sector banks (PSBs) have always dominated the banking scenario in India with their economic, national as well as social objectives. Since 1991, banks in the public sector accounted for as much as 91 per cent of the total assets as against their counterparts in the private sector with just 3 per cent and foreign sector with 6 per cent only (Choudhary & Tandon, 2010; Nandy, 2010; Purohit, 2012; Reserve Bank of India, 1991). Even today, PSBs rein the market with 73 per cent market share of assets and 83 per cent of branches (D. Subbrao, Former Governor, Reserve Bank of India, 2013).

But in such a stiff competition, with strong rival players, economies of scale and scope have to be exploited to face the competition. As a result, efficiency becomes the most critical objective. Hence, in the present article, an attempt has been made to analyze the efficiency performance of PSBs. The article is divided into in various sections. The first section, given above, is the introduction and the second section reviews the related studies available in the literature. The third section discusses objectives of the article. The fourth section outlines the econometric framework, input–output selection for DEA and methodology used in the article. The fifth section reports the empirical results. Finally, the sixth section concludes the study.

Relevant Literature Review

An adequate amount of studies which have evaluated efficiency of banks have been discussed. Table 1 shows the summary of these studies.

The review of literature highlights how a number of studies have analyzed the technical efficiency of the banks. However, only a small number of studies specifically concentrate on measuring the efficiency of PSBs (Bala & Kumar, 2011; Kumar & Gulati, 2008; Kumar, 2012; Noulas & Ketkar, 1996; Saha & Ravishankar, 2000; Tandon, 2006; Tandon, Ahuja & Tandon, 2009). Also, majority of these studies belong to an old period of time representing the pre-reform era (Bhattacharyya et al., 1997; Noulas & Ketkar, 1996; Saha & Ravishankar, 2000). Some recent studies by Kumar and Gulati (2008), Kumar (2008) and Bala and Kumar (2011) have endeavoured to measure the technical efficiency of PSBs, but the time period is limited to 1 year only. Thus, the present article specifically focuses on measuring the technical efficiency of PSBs in the most recent time period, that is, from 2007–2008 to 2011–2012. This time period not only represents the post-reformatory era but also covers the years of global financial crisis. It is believed that this study would fill the gaps of available literature.

Objective of the Study

The primary objective of the article is to analyze and evaluate technical efficiency scores of PSBs in India. The article also determines the nature of return to scale (RTS) in individual banks and identifies the leaders and laggards among banks in the public sector.

Econometric Framework

Non-parametric Approach: Data Envelopment Analysis

The estimates of efficiency are sensitive to the choice of technique (Jacobs, Smith & Street, 2006; Jaforullah & Premachandra, 2003). Both parametric and non-parametric approaches have been used to study the efficiency of banks. Parametric approach uses pre-specified functional forms, such as ‘the translog production function’ or ‘Cobb–Douglas production function’, that endeavour to bear a resemblance to the actual production process as closely as possible (Coelli, Rao & Battese, 1998). The parametric approach includes production, cost, profit and the revenue function as alternative approaches of estimating efficiency (Ajibefun, 2008). Parametric approach requires the definition of a specific functional form for the technology and for the inefficiency error term, while the non-parametric approach requires no specific functional form and chances of measurement errors here are less. The nonparametric approach has the natural advantage of eliminating the effects of all productive and scale inefficiencies prior to calculating scale economies. The non-parametric approach is less restrictive in nature as it does not require unwarranted functional form on the structure of the production technology (Avkiran, 1999; Rebelo & Mendes, 2000). The most frequently used non-parametric approach is the data envelopment analysis (DEA) as it allows diagnostics of the causes of inefficiencies.

Table 1.

Summary of Studies on the Performance Evaluation of Banks

Author/Year	Sample	Time Period	Input	Output	Results
Tandon (2006)	Public sector banks (PSBs)	From 2003 to 2008	Interest expenses Operating expenses	Business	Punjab National Bank was the most efficient bank with respect to interest expense as input. Oriental Bank of Commerce (OBC) was considered as the most efficient in terms of operating expenses as input and business as output.
Sanusi, Kusari, Wahab and Hamzah (2007)	Malaysian and Indonesian commercial banks	2001–2007	Non-interest expense Total deposits	Non-interest income Total loans	The overall technical efficiency (OTE) of the banking industry in Malaysia and Indonesia had improved during the study period from 2001 to 2007. Overall there was no striking difference in technical efficiency of Malaysian and Indonesian commercial banks.
Kumar and Gulati (2008)	27 PSBs	2004–2005	Physical capital Labour Loanable funds	Net interest income Non-interest income	The level of OTE in Indian public sector banking industry was around 88.5%. 7 PSBs scored OTE score of 1. The efficient banks included State Bank of Bikaner and Jaipur, State Bank of Mysore, State Bank of Patiala, State Bank of Travancore, Andhra Bank, Corporation Bank and Punjab and Sind Bank. OTE scores among the inefficient banks ranged from 0.632 for Bank of India (BOI) to 0.974 for Indian Overseas Bank.
Chansarn (2008)	13 Thai commercial banks	2003–2006	Operation Approach Interest expenses Labour-related expenses Capital-related expenses Intermediation Approach Total deposits Total expense	Interest and dividend incomes Non-interest incomes Total loans Net investments	For operation approach, the average efficiency of Thai commercial banks during 2003–2006 ranged from 0.9106 to 0.9720 which was considered to be very high and stable. For intermediation approach, the average efficiency of Thai commercial banks ranged from 0.7120 to 0.8693. Overall, Thai commercial banks were more efficient in operation approach than in intermediation approach.
Gupta, Doshit and Chinubhai (2008)	Indian banking sector	1999–2003	Interest expenses Operating expenses	Interest income Fee-based income Investment income	Average efficiency of all banks had increased by about 2.4% in the last 5 years. State Bank group was 2.5% more efficient than private banks and private banks were 5% more efficient than the other nationalized banks.
Kumar (2008)	27 PSBs	2005	Spread Non-interest income	Physical capital Labour Loanable funds	The mean technical efficiency for the industry was 88.5%. PSBs could produce 1.13 times as much output from the same inputs. Banks affiliated with State Bank of India (SBI) Group outperformed the banks belonging to the Nationalized Banks Group. Efficiency–profitability matrix revealed that 13 banks fell in lucky and underdog quadrants and had efficiency score below the average efficiency of industry.
Tahir, Bakar and Haron (2009)	Malaysian commercial banks	2000–2006	Total deposits Total overhead expenses	Total earning assets	The share of scale-efficient banks (CRS) was small. The share of scale-efficient domestic banks had declined over the 6-year period from 29% in 2000 to 22% in 2006. Majority of foreign banks were operating in the region of decreasing returns to scale.
Tandon et al. (2009)	19 PSBs	From 2003 to 2008	Interest expenses Operating expenses	Business	Punjab National Bank scored the maximum with respect to interest expenses as input and OBC had minimum efficiency. Among 19 PSBs, the least efficient bank in terms of interest expenses and operating expenses as input and business as output was Punjab and Sind Bank.
Bala and Kumar (2011)	27 PSBs	2008–2009	Physical capital Labour Loanable funds	Interest spread Non-interest income	PSBs had high level of efficiency as reflected by the mean efficiency score of 0.890. TE scores among the inefficient banks ranged from 0.553 for Central Bank of India (CBI) to 0.999 for State Bank of Indore.
Joshi and Bhalerao (2011)	Indian commercial banks	From 2008 to 2010	Deposits Interest expenses Operating expenses Assets	Advances and loans Investments Net interest income Non-interest income	In 2008, Punjab National Bank, Jammu & Kashmir Bank, Federal Bank and Bank of Baroda illustrated scores ranging from 0.978 to 1.025. Overall, 2010 was comparatively an efficient year for most of the banks.
Gulati (2011)	Indian domestic banks	2006–2007	Physical capital Labour Loanable funds	Net-Interest Income Non-interest income	Out of 51 sample banks, only 9 banks had been found to be overall technically efficient with OTE score equal to 1. 14 banks had been identified as relatively efficient under variable return to scale (VRS) assumption since they had attained pure technical efficiency (PTE) score equal to 1. ICICI Bank dominated the whole sample with the super-efficiency score equal to 1.66 and ranked at the top position.
Rajput and Handa (2011)	Indian banks	2004–2009	Capital Labour	Deposits Loans	On the basis of 2009 performance, Kotak Mahindra, ICICI and Yes Bank had performed well.
Yasmeen (2011)	35 Bangladesh private and public commercial banks	2003–2007	Interest expense Non-interest expenses	Income from financing or net interest income Non-financing-related income or non-interest income	The private commercial banks exhibited high efficiency in all years. Most of the banks that recorded low technical efficiency indicated that scale inefficiencies were the prime source of their technical inefficiencies.
Prabhakar, Sheriff and Nagadevara (2012)	79 Indian banks	From 2008 to 2010	Ratio of deposits to total liabilities Cost of deposits Ratio of intermediation cost to total assets Ratio of wage bills to total expense Ratio of burden to interest income	Credit + investment - deposit ratio Ratio of net interest income to total assets Return on equity Ratio of non-interest income to total assets Ratio of net NPA to net advances Return on advances adjusted to cost of funds Return on investments adjusted to cost of funds	In 2010, the efficiency of public sector banks varied from 0.55 to 1 with a mean efficiency score of 0.74. The efficiency of private sector banks varied from 0.47 to 1 with a mean efficiency index of 0.77. For foreign sector banks, the efficiency index varied from 0.46 to 1 with a mean efficiency of 0.90. Overall, the study showed that the mean efficiency index for PSBs was 0.81, 0.72 and 0.64 in 2008, 2009 and 2010, respectively. The index scores of private sector banks were 0.76, 0.74 and 0.69 and for foreign sector banks, the scores were 0.94, 0.90 and 0.90 in 2008, 2009 and 2010, respectively.
Chhikara and Bhatia (2012)	Foreign banks operating in India	2010–2011	Deposits Capital Reserves Interest expanded	Loans Investment Interest income Other income	The mean score of OTE for all the banks was found to be 0.910. PTE had a mean score of 0.938 for all the banks. According to the scale efficiency (SE), mean score for all the banks taken together was 0.969.
Karimzadeh (2012)	8 Indian commercial banks	2000–2010	Fixed assets Deposits Number of employees	Loans Investments	Average efficiency score was 0.936 with CRS, and 0.985 with VRS. According to technical efficiency (CRS), BOI and SBI showed better performance and were the most efficient banks and CBI had the lowest efficiency.
Kumar (2012)	Indian PSBs	From 1992–1993 to 2007–2008	Physical capital Labour Loanable funds	Net interest income Non-interest income	The average level of cost-efficiency (inefficiency) in Indian public sector banking industry was 79.6% (25.6%). The cost-efficiency of Indian public sector banking industry rose at a modest rate of 0.868% per annum over the entire study period. Overall, the majority of PSBs belonging to the Nationalized Banks Group had recorded a significant improvement in cost, technical and allocative efficiencies since 1998–1999.
Zeitun and Benjelloun (2013)	12 Jordanian banks	2005–2010	Model A Deposits Equity capital Other assets Model B Deposits Equity Fixed assets Model C Deposits Equity Other assets Fixed assets	Net income before tax Net income Loans Net interest Other earning assets	The mean efficiency score of Jordanian banks was 54% as per Model A, using CRS, while it was 67% using VRS. Model B had mean efficiency score of 68% in CRS and 80% using VRS. Model C got the highest average 89% using VRS and 82% using CSR. The bank-wise result showed that under Model A, out of 12 banks only 2 banks, Jordan Kuwait Bank (JKB) and Dubai Islamic Bank (DIB), were on the frontier. Under Model B, JKB was on the frontier in years 2005, 2006, 2008, 2009 and 2010, while DIB was efficient only in 2005, 2007 and 2009 based on CRS.

Source: Author’s Own.

Farrell (1957) used the economic concept of production frontier and production possibility set to define technical and allocative efficiency. Later on, Charnes, Cooper and Rhodes (1978) (CCR model) extended Farrell’s (1957) idea into non-parametric methodology; popularly known as DEA. It is a linear programming-based technique employed for assessing the relative performance of a set of firms that uses a variety of inputs to produce a variety of outputs. The DEA has the capacity to consider multiple inputs and outputs to measure efficiency against the best-observed performance known as ‘the Efficiency Frontier’. The DEA identifies efficiencies of all firms relative to the best practice firm in the sample. A firm in DEA is known as decision-making unit (DMU). The main aim of DEA is to recognize which firms are operating on the production frontier having an efficiency score of one and which firms perform below the frontier that have a score between zero and one. The firm having a score of one is the most efficient firm, while the firm having a score between zero and one is less efficient.

The overall technical efficiency is decomposed into two components: pure technical efficiency and scale efficiency (SE). The overall technical efficiency assumes the constant return to scale (CRS), whereas pure technical efficiency assumes the variable return to scale (VRS). A CRS implies that a change in the amounts of the inputs leads to a similar change in the amounts of the outputs. The VRS determines how the scale of operation slows down a firm’s performance. The VRS includes both increasing returns to scale (IRS) and decreasing returns to scale (DRS). The IRS exists when the proportional increase in all inputs results in a more than proportional increase in outputs. The DRS exists when an increase in inputs results in a lower increase in outputs.

Input–Output Selection for DEA

The selection of input and output variables is essential for the successful application of DEA. Different authors have suggested different approaches for selecting input and output of banks. The two main approaches include the production approach and the intermediate approach. The production approach views banks as using the purchased inputs in terms of the operating cost and interest expenses to produce deposits, loans and advances. The intermediate approach views the banks as using deposits together with purchased inputs and operating costs to produce loans and advances. To calculate the efficiency of banks, researchers have followed both production approach (Bhattacharyya et al., 1997; Chansarn, 2008; Ketkar & Ketkar, 2008; Rammohan & Ray, 2004; Saha & Ravishankar, 2000; Jemric & Vujcic, 2002) and intermediate approach (Ataullah & Hang, 2006; Chansarn, 2008; Das, Nag & Ray, 2005; Karimzadeh, 2012; Ketkar & Ketkar, 2008; Sahoo, Sengupta & Mandal, 2007; Sanusi et al., 2007; Varadi et al., 2006; Jemric & Vujcic, (2002); Yue, 1992). However, Favero and Papi (1995) and Berger and Humphrey (1997) have pointed out that intermediation approach is the most appropriate for banks because most activities of banks involve converting huge deposits and funds into loans and financial investments. Thus, this article employs intermediation approach and takes deposits, borrowings, interest expenses and operating expenses as inputs and investments, advances, interest income and non-interest income as outputs.

Data Base and Research Methodology

The article analyzes the efficiency of currently operating PSBs in India, that is, 26 banks. State Bank of Saurashtra and State Bank of Indore are excluded from the sample for the reason that these banks were merged with the State Bank of India. The sample size is considered to be satisfactory as it represents more than twice the sum of inputs and outputs, which is a precondition for the application of DEA. The technical efficiency performance of PSBs has been analyzed over 5 years from 2007–2008 to 2011–2012. It is the most recent period for which data are available. In addition, these years represent the most critical time period for banks as it covers the years of global financial crisis as well. The study is based on secondary data. The data have been collected from the banks’ annual reports, websites of Reserve Bank of India (RBI) and Reports on Trend and Progress in Banking from 2007–2008 to 2011–2012. The study uses Data Envelopment Analysis Program (DEAP 2.1) developed by Coelli (1996) to calculate the efficiency of PSBs in India.

Empirical Results and Findings

Individual Bank Efficiency

In this section, overall technical efficiency (CRS), pure technical efficiency (VRS) and SE are obtained by employing DEA. The estimated efficiencies are reported in Table 2.

Table 2.

Efficiency Scores of Public Sector Banks in India

	Banks	2007–2008				2008–2009				2009–2010				2010–2011				2011–2012
	Banks		CRS	VRS	SE	Return	CRS	VRS	SE	Return	CRS	VRS	SE	Return	CRS	VRS	SE	Return	CRS	VRS	SE	Return
1	Allahabad Bank	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	0.975	1.000	0.975	DRS	0.969	0.970	1.000	–
2	Andhra Bank	1.000	1.000	1.000	–	0.983	0.984	0.999	DRS	0.971	0.982	0.989	IRS	1.000	1.000	1.000	–	0.994	1.000	0.994	IRS
3	Bank of Baroda	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–
4	Bank of India	1.000	1.000	1.000	–	1.000	1.000	1.000	–	0.980	0.981	0.999	DRS	0.985	1.000	0.985	DRS	0.960	0.964	0.995	DRS
5	Bank of Maharashtra	1.000	1.000	1.000	–	1.000	1.000	1.000	–	0.938	0.956	0.981	IRS	0.988	1.000	0.988	IRS	0.925	0.974	0.950	IRS
6	Canara Bank	1.000	1.000	1.000	–	0.988	1.000	0.988	DRS	1.000	1.000	1.000	–	1.000	1.000	1.000	–	0.979	1.000	0.979	DRS
7	Central Bank of India	1.000	1.000	1.000	–	1.000	1.000	1.000	–	0.946	0.965	0.980	DRS	0.950	0.951	0.998	DRS	0.865	0.866	0.999	IRS
8	Corporation Bank	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–
9	Dena Bank	1.000	1.000	1.000	–	0.974	1.000	0.974	IRS	0.971	1.000	0.971	IRS	0.999	1.000	0.999	IRS	0.963	1.000	0.963	IRS
10	IDBI Bank Ltd	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–
11	Indian Bank	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–
12	Indian Overseas Bank	0.996	0.996	1.000	–	1.000	1.000	1.000	–	0.980	0.987	0.993	IRS	0.990	0.991	0.999	IRS	0.927	0.928	0.999	IRS
13	Oriental Bank of Commerce	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–
14	Punjab and Sind Bank	0.979	1.000	0.979	IRS	1.000	1.000	1.000	–	0.977	1.000	0.977	IRS	0.986	1.000	0.986	IRS	1.000	1.000	1.000	–
15	Punjab National Bank	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	0.857	0.913	0.939	IRS
16	State Bank of Bikaner & Jaipur	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	0.946	1.000	0.946	IRS
17	State Bank of Hyderabad	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	0.993	0.997	0.996	IRS	1.000	1.000	1.000	–
18	State Bank of India	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–
19	State Bank of Mysore	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	0.912	1.000	0.912	IRS
20	State Bank of Patiala	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	0.959	1.000	0.959	IRS
21	State Bank of Travancore	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	0.911	0.962	0.947	IRS
22	Syndicate Bank	0.996	1.000	0.996	DRS	0.994	1.000	0.994	IRS	1.000	1.000	1.000	–	1.000	1.000	1.000	–	0.931	0.934	0.996	IRS
23	UCO Bank	0.965	0.977	0.988	DRS	0.971	0.973	0.998	IRS	1.000	1.000	1.000	–	1.000	1.000	1.000	–	0.976	0.977	0.999	DRS
24	Union Bank of India	1.000	1.000	1.000	–	0.988	0.995	0.993	DRS	0.998	1.000	0.998	DRS	0.979	0.990	0.989	DRS	0.953	0.954	0.999	IRS
25	United Bank of India	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	1.000	1.000	1.000	–	0.958	0.985	0.973	IRS
26	Vijaya Bank	1.000	1.000	1.000	–	1.000	1.000	1.000	–	0.983	0.997	0.986	IRS	1.000	1.000	1.000	–	0.940	1.000	0.940	IRS
	Mean	0.998	0.999	0.999		0.996	0.998	0.998		0.990	0.995	0.995		0.994	0.997	0.997		0.959	0.978	0.980

Source: Author’s Own.

Table 2 indicates that there has been a change in the efficiency patterns of banks over 5 years from 2007–2008 to 2011–2012. In 2007–2008, the most efficient banks comprized 22 banks. These banks operated with full pure technical and scale efficiencies as they have an efficiency score of 1. Only four banks, that is, Indian Overseas Bank, Punjab and Sind Bank, Syndicate Bank and UCO bank have an efficiency score of less than 1 which makes them less efficient. In 2008–2009, four banks, namely, Andhra Bank, Canara Bank, Dena Bank and Union Bank of India, were dropped out of the record of efficient banks, whereas Indian Overseas Bank and Punjab and Sind Bank entered into the catalogue of efficient banks. Overall, 20 banks remained fully efficient in 2008–2009. Further, the list of fully efficient banks went down to 17 banks in 2009–2010. Bank of India, Bank of Maharashtra, Central Bank of India, Indian Overseas Bank, Punjab and Sind Bank and Vijaya Bank became inefficient in the year 2009–2010. Canara Bank, Syndicate Bank and UCO Bank turned out to be efficient. On the whole, in 2010–2011, the number of efficient banks remained the same as in 2009–2010 as Andhra bank and Vijaya Bank changed into efficient banks, while Allahabad Bank and State Bank of Hyderabad turned out to be inefficient banks. In 2011–2012, only 7 banks remained fully efficient while the remaining 19 banks had an efficiency score varying from a low of 0.857 to a high of 0.999.

On the whole, Bank of Baroda, Corporation Bank, IDBI Bank, Indian Bank, Oriental Bank of Commerce and State Bank of India exhibit a high efficiency score of 1 during all years. These banks are considered as fully efficient as the input–output combination of these banks lies on the Efficient Frontier for all these years. Many of these banks have expanded their scope and operational efficiency by introducing electronic banking, Internet banking, mobile banking, credit cards, automatic teller machines (ATMs), electronic fund transfer (EFTs), real-time gross settlement (RTGS) and national electronic funds transfers (NEFTs).

The mean CRS efficiency score of PSBs shows an evident decline from 0.998 in 2007–2008 to 0.959 in 2011–2012. The mean VRS efficiency score of PSBs also depicts a decline from 0.999 in 2007–2008 to 0.978 in 2011–2012. Even the fully efficient number of banks declined from 22 to an alarming low of 7. This decline in the number of fully efficient banks is perhaps due to deceleration in the balance sheet of PSBs in 2011–2012, in terms of asset quality. This reduced the deposit mobilization by banks and increased the cost of deposits, as well as the cost of borrowings, resulting in lower spread and inefficiency. Moreover, the trimmed down performance of Indian banks, to some extent, has been attributed to the high inflation and muted growth performance of Indian economy as a whole and the fragile recovery of the Indian financial market from the ripples of global recession in 2011–2012 (Reports on Trend and Progress of Banking in India 2011–2012, Reserve Bank of India, 2012).

However, the individual scores of banks suggest an optimistic view as all the PSBs have more or less similar efficiency scores, that is, the efficiency score for majority of banks is higher than 0.900. These efficiency scores also suggest that a greater part of inefficiency among PSBs is attributed to scale inefficiency. Scale inefficiency cautions that banks are not operating on the optimum scale. They need to expand their business not only by opening new branches but also by increasing their customer base by improving the quality of services and by effective customer relation management to achieve economies of scale. The rest of the inefficiency among PSBs is attributed to pure technical inefficiency. Pure technical inefficiency depicts that management of these banks needs to gear up its operational abilities. Perhaps, it needs to motivate its personnel, improve its processes and upgrade its technologies to achieve operational synergy.

Analyzing Return to Scale

As per the results generated by DEA (Table 2), the nature of RTS may vary from CRS to IRS or DRS. The DEA also helps to determine whether a DMU is operating at a wrong scale, that is, either operating at DRS or IRS. The DRS indicates that DMU is operating at a scale that is too large which portrays that a percentage increase in inputs of that DMU produces a less than proportional increase in outputs. DMUs operating on DRS can improve their performance by compressing their scale of operations. On the other hand, IRS depicts that DMU is operating at a scale that is too small, which means that the percentage increase in inputs of DMU produces a more than proportional increase in outputs. The DMU with IRS lies below the optimum scale and would improve their operations by mounting their scale of operations. Table 3 provides a synoptic view of the number of PSBs operating at different RTSs.

Table 3.

Number of Public Sector Banks at Different Return to Scale

Year	IRS	DRS	CRS	Total
2007–2008	1 (4)	2 (8)	23 (88)	26
2008–2009	3 (12)	3 (12)	20 (76)	26
2009–2010	6 (23)	3 (11)	17 (66)	26
2010–2011	5 (19)	4 (15)	17 (66)	26
2011–2012	14 (54)	3 (11)	9 (35)	26

Source: Authors’ Own.

Note: Figures in parentheses are percentages.

Table 4.

Number of Public Sector Banks Identified as Leaders and Laggards

Class of Bank	Leaders			Moderate Performers			Laggards
Year	CRS	VRS	SE	CRS	VRS	SE	CRS	VRS	SE
2007–2008	22	24	23	0	0	0	4	2	3
2008–2009	20	23	20	0	0	2	6	3	4
2009–2010	17	20	17	1	1	2	8	5	7
2010–2011	17	22	17	1	1	3	8	3	6
2011–2012	8	15	9	7	1	7	11	10	10

Source: Authors’ Own.

Table 3 shows that the number of PSBs operating at IRS increased tremendously from 1 (4 per cent) to 14 (54 per cent) from 2007–2008 to 2011–2012. The number of banks operating at DRS remained almost the same with a minor variation. As a result, the number of banks operating at CRS fell drastically from 23 (88 per cent) in 2007–2008 to 9 (35 per cent) in 2011–2012. This implies that majority of banks need to enlarge their scale of operations. Scale inefficiency seems to be a very gross cause of poor performance of PSBs in India.

Identification of Leaders and Laggards

On the basis of the mean efficiency scores generated by DEA, PSBs have been divided into three categories: (i) banks with an efficiency score of 1 are considered as leaders, (ii) banks which have an efficiency score less than 1 but greater than mean efficiency score are rated as moderate performers and (iii) banks with a mean efficiency score of less than average are considered as laggards. Table 4 presents a bird’s eye view of the number of banks falling in each category.

It is observed that there is a reduction in leaders from 22 to 8 (CRS), 24 to 15 (VRS) and 23 to 9 (SE) during the time period under study, while laggards have increased from 4 to 11 (CRS), 2 to 10 (VRS) and 3 to 10 (SE) from 2007–2008 to 2011–2012. This decline in leaders and increase in laggards seems to be due to scale inefficiencies. These banks are slow in technological upgradations and hence are unable to capture the economies of operating on a large scale. Also, PSBs are mandatorily required to open branches in rural and semi-urban areas. Further, they are compulsorily bound to advance 40 per cent towards Priority Sector Lending (PSL). This limits their opportunity to generate earning assets and results in lesser output with the same input. This gradually affects their efficiency scores and makes them lag behind their counterparts.

Conclusion

The snapshot of results is as follows:

The efficiency scores of majority of banks, that is, approximately 73 per cent have declined during the time period from 2007–2008 to 2011–2012. As a result, only 7 banks out of the 26 banks operate at full efficiency.

From the 23 banks operating at CRS in 2007–2008, the number fell to just 9 banks in 2011–2012, suggesting that PSBs were not able to maintain their input–output synchronization.

The proportion of laggards among PSBs increased almost three to five times, while that of leaders was reduced with the same percentage over a total span of 5 years.

There is an evident effect of slowdown in Indian economy on the efficiency of PSBs. In spite of RBI’s traditional banking policies, the effect of macro-events could not be controlled fully. In a globalized environment, where financial system is integrated and interdependent, the ripples of turmoil in one country definitely travel to another country. US recession affected the sentiments and faith of people in banking globally. Money is the inventory of banking business and it has a huge opportunity cost. No doubt, macro-events are not fully controllable, still at the micro-level PSBs should make an endeavour to optimize their scale of operations by adopting technological upgradations and bringing managerial synergies.

Footnotes

Acknowledgements

The authors are grateful to the anonymous referees of the journal for their extremely useful suggestions to improve the quality of this article. Usual disclaimers apply.

References

Ajibefun

(2008). An evaluation of parametric and non-parametric methods of technical efficiency measurement: Application to small scale food crop production in Nigeria. Journal of Agriculture & Social Sciences, 4(3), 95–100.

Ataullah

, & Hang

(2006). Economic reforms and bank efficiency in developing countries: The case of the Indian banking industry. Applied Financial Economics, 16(9), 653–663.

Avkiran

N.K.

(1999). An application reference for data envelopment analysis in branch banking: Helping the novice researcher. International Journal of Bank Marketing, 17(5), 206–220.

Bala

, & Kumar

(2011). How efficient are the public sector banks in India? An empirical investigation. Sri Krishna International Research & Educational Consortium, 1(3), 39–62.

Berger

A.N.

, & Humphrey

(1997). Efficiency of financial institutions: International survey and directions for future research. European Journal of Operational Research, 98(2), 175–212.

Bhattacharyya

, Lovell

C.K.

, & Sahay

(1997). The impact of liberalization on the productive efficiency of Indian commercial banks. European Journal of operational research, 98(2), 332–345.

Chansarn

(2008). The relative efficiency of commercial banks in Thailand: DEA approach. International Research Journal of Finance and Economics, 18(3), 53–68.

Charnes

, Cooper

W.W.

, & Rhodes

(1978). Measuring the efficiency of decision making units. European Journal of Operational Research, 2(4), 429–444.

Chhikara

, & Bhatia

(2012). Measurement of efficiency of foreign banks in India through data envelopment analysis (DEA). International Journal of Management Sciences, 1(3), 40–50.

10.

Choudhary

, & Tandon

(2010). Performance evaluation of public sector banks in India. Asia Pacific Journal of Research in Business Management, 1(1), 1–17.

11.

Coelli

T. J.

(1996). Measurement and sources of technical efficiency in Australian coal-fired electricity generation. Centre for Efficiency and Productivity, Analysis (CEPA), Department of Econometrics, University of New England, Armidale, Australia’,(WP 1/96).

12.

Coelli

, Rao

, & Battese

(1998). An introduction to efficiency and productivity analysis. USA: Kluwer Academic Publishers.

13.

Das

, Nag

, & Ray

S.C.

(2005). Liberalisation, ownership and efficiency in Indian banking: A nonparametric analysis. Economic and Political Weekly, 40(12), 1190–1197.

14.

Farrell

M.J.

(1957). The measurement of productivity and efficiency. Journal of the Royal Statistical Society (Series A—General), 120(2), 253–281.

15.

Favero

C.A.

, & Papi

(1995). Technical efficiency and scale efficiency in the Italian banking sector: A non-parametric approach. Applied Economics, 27(4), 385–396.

16.

Gulati

(2011). Estimation of technical, pure technical and scale efficiencies of Indian banks: An analysis from cross-sectional perspective. The 13th Annual Conference on Money and Finance in the Indian Economy, Indira Gandhi Institute of Development Research, Mumbai.

17.

Gupta

O.K.

, Doshit

, & Chinubhai

(2008). Dynamics of productive efficiency of Indian banks. International Journal of Operations Research, 5(2), 78–90.

18.

Jacobs

, Smith

P.C.

, & Street

(2006). Measuring efficiency in health care: Analytical techniques and health policy. United Kingdom: Cambridge University Press.

19.

Jaforullah

and Premachandra

2003, Sensitivity of technical efficiency estimates to estimation approaches: an investigation using New Zealand dairy Industry Data. Economics Dusicussion paper No. 306, University of Otago. Retrieved 12 November 2013, from http://www.business.otago.ac.nz/econ/research/discussionpapers/DP0306.pdf06.2005_11.2004

20.

Jemric

, & Vujcic

(2002). Efficiency of banks in Croatia: A DEA approach. Comparative Economic Studies, 44(2-3), 169–193.

21.

Joshi

P. V.

, & Bhalerao

J. V.

(2011). Efficiency evaluation of banking sector in India based on data envelopment analysis. Indian Journal of Commerce & Management Studies ISSN, 2229, 5674.

22.

Karimzadeh

(2012). Efficiency analysis by using the data envelopment analysis model: Evidence from Indian banks. International Journal of Latest trends in Finance & Economics Sciences, 2(3), 228–237.

23.

Ketkar

K.W.

, & Ketkar

S.L.

(2008). Performance and profitability of Indian banks in the post liberalization period. Presented at the 2008 World Congress on National Accounts and Economic Performance Measures for Nations, Washington, DC.

24.

Kumar

(2008). An analysis of efficiency and profitability relationship in Indian public sector banks. Global Business Review, 9(1), 115–129.

25.

Kumar

(2013). Banking reforms and the evolution of cost efficiency in Indian public sector banks. Economic Change and Restructuring, 46(2), 143–182.

26.

Kumar

, & Gulati

(2008). An examination of technical, pure technical, and scale efficiencies in Indian public sector banks using data envelopment analysis. Eurasian Journal of Business and Economics, 1(2), 33–69.

27.

Nandy

(2010). Banking sector reforms in India and performance evaluation of commercial banks. Retrieved 12 November 2013, from http://www.bookpump.com/dps/pdf-b/9423510b.pdf

28.

Noulas

A.G.

, & Ketkar

K.W.

(1996). Technical and scale efficiencies in the Indian banking sector. International Journal of Development Banking, 14(1), 19–27.

29.

Prabhakar

A.K.

, Sheriff

S.F.

, & Nagadevara

(2012). Relative efficiency of banks in India—evaluation and policies for improvement. Ninth AIMS International Conference on Management, International Institute of Management, Pune.

30.

Purohit

2012, A Critical Analysis of Financial Reforms in banking sector in post liberalization period: With Respect to Public and Private Sector Bank. A Thesis submitted to Mohanlal Sukhadia University, Udaipur.

31.

Rajput

, & Handa

(2011). Banking efficiency: Application of data envelopment approach (DEA). International Journal of Research in Commerce and Management, 2(9), 65–70.

32.

Rammohan

T.T.

, & Ray

(2004). Comparing performance of public and private sector banks: A revenue maximization efficiency approach. Economic and Political Weekly, 39(12), 1271–1275.

33.

Rebelo

, & Mendes

(2000). Malmquist indices of productivity change in Portuguese banking: The deregulation period. International Advances in Economic Research, 6(3), 531–543.

34.

Reserve Bank of India. (1991). Report on Trend and Progress of Banking, 1991–92.

35.

Reserve Bank of India. (2012). Report on Trend and Progress of Banking, 2011–12.

36.

Saha

, & Ravisankar

T.S.

(2000). Rating of Indian commercial banks: A DEA approach. European Journal of Operational Research, 124(1), 187–203.

37.

Sahoo

B.K.

, Sengupta

J.K.

, & Mandal

(2007). Productive performance evaluation of the banking sector in India using data envelopment analysis. Retrieved 9 November 2013, from http://papers.ssrn.com/sol3/papers.cfm/abstract_id=956812

38.

Sanusi

, Kusari

, Wahab

, & Hamzah

(2007). Technical efficiency of Malaysian and Indonesia commercial banks: A data envelopment analysis (DEA) approach. Retrieved 20 October 2013, from http://www.academia.edu/1540511/Technical_Efficiency_of_Malaysian_and_Indonesian_Commercial_Banks_A_Data_Envelopment_Analysis_DEA_Approach

39.

Subbarao

(2013, August 13). Banking structure in India looking ahead by looking back. Speaking Notes at the FICCI-IBA Annual Banking Conference, Mumbai.

40.

Tahir

, Bakar

, & Haron

(2009). Evaluating efficiency of Malaysian banks using data envelopment analysis. International Journal of Business and Management, 4(8), 96–106.

41.

Tandon

(2006). Performance variances & efficiency parameters of the Indian public sector banks—a suggestive (nonparametric) DEA model. IBS Conference, Hyderabad.

42.

Tandon

, Ahuja

, & Tandon

(2009). Relative performance of banks. The Indian Banker, 4(7), 39–45.

43.

Varadi

V. K.

, Mavaluri

P. K.

, & Boppana

(2006). Measurement of efficiency of banks in India, Munich Personal RePEc Archive. Retrieved 14 July 2013, from http://mpra.ub.uni-muenchen.de/17350/MPRAPaperNo.17350

44.

Yasmeen

(2011). Technical efficiency in the Bangladeshi banking Industry: A non parametric analysis. The 2011 Barcelona European Academic Conference Barcelona, Spain.

45.

Yue

(1992). Data envelopment analysis and commercial bank performance: A primer with applications to Missouri banks. Federal Reserve Bank of St. Louis Review, 74(1), 31–45.

46.

Zeitun

, & Benjelloun

(2013). The efficiency of banks and the financial crisis in a developing economy: The case of Jordan. Journal of Finance, Accounting and Management, 4(1), 1–20.