Earnings Quality of Commercial Banks in the Post- liberalized Era: A Multivariate Analysis

Transcription

1 ABSTRACT Earnings Quality of Commercial Banks in the Post- liberalized Era: A Multivariate Analysis Dr. O C Aloysius Associate Professor of Commerce Government College, Kattappana, Kerala - India The banking sector reforms introduced in the 1990 s have changed the face of the banking sector in India. New private banks were started functioning in India with new technology after a long period of time and creates tougher competition between the existing players and the new generation banks. Profitability is one of the key indicator used to test the performance of a bank. This can be achieved only by increasing the income and operational performance. In this paper, an attempt is done to measure and compare the earnings quality of the top eight banks six selected public sector banks and two from the new generation private sector banks, bank-wise and bank sector-wise by considering three core areas viz., operational performance, profitability and source of income. In this paper, MANCOVA technique is used to find out the difference between banks and bank sectors. It is found that HDFC Bank Ltd and SBT ranks first and second respectively in terms of overall earnings quality performance and the difference in the performance of new generation private sector banks and the public sector banks are highly relevant. In terms of operational performance, Canara Bank is the best bank and HDFC Bank Ltd is the best bank in terms of profitability and source of earnings. Key words: Earnings quality, Profitability, Operational performance, I. INTRODUCTION The banking industry occupies a unique place in a nation s economy by channelising money from the one who has surplus money to those persons or sectors which need it. Indian banking sector, one of the largest banking systems in the world had been passing through different phases from privatization to nationalization and again to privatization. The public sector banks enjoyed monopoly in the market and created lot of problems mainly inefficiency at all levels and lack of profitability. The reforms that took place during the 1991 brought a sea change in the banking sector. Today by the advent of economic reforms and the opening up of banking industry to private parties, new generation techno savvy banks and foreign banks were entered in the market. This creates tougher competition between the existing players and new generation banks. Public sector banks were awakened from their deep sleep by changing their marketing strategies, diversifying products and services, adopting new technology, customer service and allowing private investment. Various factors influence the performance of a bank and the earnings efficiency of a bank is of prime importance. Earnings quality and profitability reflect the sustainability and efficiency of a bank, and earning capacity of assets ensures the continuity of the profit -making process. If the assets or investments made by the bank are not yielding adequate return, they do not have a satisfactory earning capacity. Earnings quality ratios throw light on the profitability of the bank from the viewpoint of owners and its operating efficiency. Many studies were already done relating to the performance of public sector banks and new generation banks. But none of the studies measures and compare earnings efficiency of commercial banks in India by using the MANCOVA technique by controlling the effect of the covariate, time. So in this paper an attempt is made to analyse and compare the earnings quality of top banks in India. IRJBM ( ) Volume No VII September Issue 9 Page 18

2 II. OBJECTIVES, RESEARCH METHODOLOGY AND DATA BASE 2.1 Objectives 1. To analyze the earnings quality performance of the largest public sector banks and new generation private sector banks on three dimensions, bank-wise and bank sector-wise. 2. To make a comparative analysis of Earnings Quality Performance of public sector banks and new generation private sector banks. 2.2 Hypotheses Hypothesis I H0: There is no difference in the means of earnings quality ratios due to trend over time between banks, and bank sectors (public sector banks and new generation private sector banks). Ha: There is significant difference in the means of earnings quality ratios due to trend over time between banks, and bank sectors (public sector banks and new generation private sector banks). Hypothesis II H0: There is no difference in the earnings quality ratios between public sector banks and new generation private sector banks. Ha: There is significant difference in the earnings quality ratios between public sector banks and new generation private sector banks. 2.3 Variables used in the study Although there are various measures used to calculate the earnings performance of a bank, the following eight ratios are used for studying earnings quality. Sl.No Earnings Quality Ratios Criteria of analysis 1 Operating profit to working fund (OP Higher ratio is Operational to WF) performance 2 Operating expenses to total income (OE to TI) Lower ratio is 3 Burden to interest income (B to Lower ratio is II) 4 Percentage growth in net profit Higher ratio is (GINP) Profitability 5 Net profit to average assets (ROA) Higher ratio is 6 Net interest margin (spread) to Total assets (NIM) Higher ratio is 7 Interest income to working fund (II to Higher ratio is Source of earnings WF) 8 Non -interest income to total income (NII to TI) Higher ratio is 2.4 Methodology adopted The research design adopted for the study is analytical in nature. After reviewing the available literature and assessing the gap that exist in this field, the researcher identified eight indicators categorized under three heads to assess the earnings quality of commercial banks The universe - The universe of the study consists of all banks (7 now six in number) in the state bank group (SBG), all banks excluding IDBI bank (19) in the other nationalized banks (ONB) and 7 new generation private sector banks (NGPB) operating in India. Totally eight banks were selected for the study. Sample banks were selected from each group based on the basis of asset size and volume of business as on The banks having large volume of business and asset size in their group is taken except SBT. IRJBM ( ) Volume No VII September Issue 9 Page 19

3 2.4.2 Sample banks 1. State bank group - SBI and SBT 2. Other nationalize banks - PNB, BOB, BOI and CB 3. New generation private sector banks - ICICI Bank Ltd and HDFC Bank Ltd Period of study - The study is confined to the post reform era i.e., after the introduction of the new economic policies by the government in The data for the ten year period from the financial year to was used Type of data used - The study relies mainly on secondary data published by various Institutions and organizations concerned with commercial banks. The publications of Reserve Bank of India, IBA and the Annual Reports of respective banks are the main sources of data Analysis of data - Ratios and averages are used for analysis. Multivariate analysis of co-variance (MANCOVA) technique is used to find out the variations in the mean ratios between banks, bank groups and bank sectors. The degree of variation in each ratio between banks and bank sectors are analyzed on the basis of the value of partial eta squared. SPSS soft ware is used for the final analysis. III - RESULTS AND DISCUSSIONS 3.1 Analysis based on Average Ratios Table 1 Average Earnings quality ratios of banks, bank groups and bank sectors Sl. EARNINGS QUALITY RATIOS No O.P to NP to AA II to NII BANKS OE to TI B to II GINP NIM WF (ROA) WF to T.I 1 SBI SBT SBG(AV) PNB BOB BOI CB ONB(AV) PSB (AV.) ICICI HDFC NGPB (AV) Source: Compiled and calculated from the Annual Reports of banks Bank-wise analysis Profitability Only with reasonable profit a bank can survive and do its business smoothly. So, profitability ratios are a clear indication of strength and efficiency. Three ratios are used, namely, percentage growth in net profit, net profit to average assets (ROA), net interest margin or spread to total assets (NIM). In terms of net profit to average assets (ROA), the most important profitability measure which reflects the profitability position, HDFC Bank Ltd ranks first, followed by BOB. In terms of percentage growth in IRJBM ( ) Volume No VII September Issue 9 Page 20

4 net profit, ICICI Bank Ltd shows the highest growth rate (63.39%), followed by Bank of India (41.99%). Further in terms of net interest margin (spread), i.e., the difference between interest earnings and interest payments as a percentage to total assets, HDFC Bank Ltd has the highest margin (3.92), followed by PNB (3.42) Source of earnings The source of earning income is very much important for a bank as its main function is to lend money to the needy people. A bank has two sources of income, viz., traditional (interest income) as well as newer source (fee-based income). In terms of II to WF, SBT has the highest ratio, followed by PNB. In the mean ratio of Non-interest income to total income, ICICI Bank Ltd. has the highest rank, followed by HDFC Bank Ltd Operational performance Three indicators, namely, operating profit to working fund (OP to WF), operating expenses to total income (OE to TI), and burden to interest income (B to II), were used to find out the operational effectiveness of a bank. Operating profit to working fund ratio reflects the operational efficiency of a bank. A bank should control its operating expenses to increase its revenue; otherwise it will cause burden on its earnings. Operating expenses include all expenses other than interest expenses. Burden means the difference between operating expenses and non- interest income. There is only slight change in all the ratios between the banks. In terms of OP to WF, HDFC Bank Ltd. stands first, followed by PNB. Regarding OE to TI, SBT has the lower ratio followed by CB. In the ratio of B to II, ICICI Bank Ltd and CB stands first and second respectively Performance of banks in the core areas of earnings quality The ranks of banks in the core areas of earnings quality is given in table 2. Table 2 Performance of banks in the core areas of earnings quality Sl No Bank Operational performance (Av. of ranks in OP2WF+OE2TI+B2II) Profitability (Av. of ranks in GINP+ NIM +ROA) Source of earnings (Av. of ranks in II to WF+ NII to T.I) 1 SBI 7.00 V 7.00 V 6.50 V 2 SBT 2.66 I 4.00 II 4.00 II 3 PNB 5.33 IV 4.30 III 4.00 II 4 BOB 5.33 IV 4.00 II 5.00 IV 5 BOI 4.00 II 5.33 IV 5.00 IV 6 CB 2.66 I 5.33 IV 4.50 III 7 ICICI 4.00 II 4.00 II 4.50 III 8 HDFC 5.00 III 2.00 I 2.50 I Overall earnings quality performance of banks Earnings quality of banks is measured by taking averages of the ranks in all ratios used in the study. Table 3 shows the ranks of selected banks. Table 3 Ranks of Banks under overall Earnings Quality Performance Sl No Bank OP to WF OE to TI B to II GINP NIM RANKS NP to AA (ROA) Position of banks IRJBM ( ) Volume No VII September Issue 9 Page 21 II to WF NII to T.I Total Av. Rank 1 SBI SBT PNB BOB

5 5 BOI CB ICICI HDFC On the basis of overall earnings performance, HDFC Bank Ltd occupies the first position followed by SBT. ICICI Bank Ltd and CB share the third position. In terms of operational performance, SBT and CB share the first rank followed by ICICI Bank Ltd. SBI occupies the last position. As profitability is considered, HDFC Bank Ltd. shows excellent performance, followed by SBT, BOB and ICICI Bank Ltd. The performance of SBI is very poor, compared to other banks. Regarding source of earnings, HDFC Bank Ltd. comes first, followed by PNB and SBT and SBI occupies the last position Bank-wise analysis- Earnings Quality Ratios and Variations among banks Due to trend over time Hypothesis 1 H0: There is no significant difference in the earnings quality ratios of banks due to trend over time a. Difference in Earnings Quality ratios of banks due to time Table 4 Earnings quality ratios -variations among banks due to time Effect Value F Hypothesis Error Sig. Partial Eta Squared Time Pillai's Trace b Wilk s Lambda b Hotelling's Trace b Roy's Largest Root b a. The statistic is an upper bound on F that yields a lower bound on the significance level. b. Exact statistic c. Design: Bank + Time Considering the differences in the means of eight earnings quality ratios due to trend over time, the multivariate analysis of covariance (MANCOVA) results, Table 4 show that there is difference in the earnings quality ratios as Pillai's Trace F value 8.109, which is significant at 5% level of significance, i.e., the P value <0.05 and the associated Partial Eta Squared is Hence we may say that there is significant difference among the earnings quality ratios of banks due to time and the null hypothesis is rejected. b. Ratio-wise difference in Earnings Quality of banks due to time While individually considering the earnings quality ratios of banks, it may be seen that the difference is found significant in three ratios only, namely, net profit to average assets (ROA), interest income to working fund (II to WF) and burden to interest income (B to II), as the P Value < 0.05.The difference is found insignificant in the ratios of OP to WF, PGINNP, NIM, OE to TI and NII to TI, because the significance value at 5% level of significance is >0.05. IRJBM ( ) Volume No VII September Issue 9 Page 22

6 c. Degree of variation in ratios Table 5 Variations in Earnings quality ratios among banks due to time Sl. No. Ratios F value Significance value Partial Eta Squared 1 OP to WF PGINNP NIM NP to AA II to WF OE to TI NII to TI B to II The degree of variation is studied on the basis of Partial Eat Squared and it is seen that II to WF seems to vary more as the associated Partial Eta Squared is 0.130, which is higher than that of all other ratios and the variation is little in the ratio of operating profit to working fund (Partial Eta Squared is 0.011) Difference in Earnings quality ratios of banks after controlling for possible variations due to trend over time a. Difference in earnings quality ratios after controlling time Hypothesis II H0: There is no significant difference in the earnings quality ratios of banks after controlling for possible variations due to trend over time Table 6 Earnings quality ratios -variations among banks after controlling for time Effect Value F Hypothesis Error IRJBM ( ) Volume No VII September Issue 9 Page 23 Sig. Partial Eta Squared Bank Pillai's Trace Wilk s Lambda Hotelling's Trace Roy's Largest Root a a. The statistic is an upper bound on F that yields a lower bound on the significance level. b. Exact statistic c. Design: Bank + Time The result of Multivariate analysis of co-variance (MANCOVA, Table 6) reveals that Pillai s Trace F value is 3.989, which is significant as the P value which is less than 0.05 and the associated Partial Eta Squared is Therefore the difference in the earnings quality performance of banks after controlling for possible changes due to trend over time is highly significant. Hence the null hypothesis is rejected. b. Ratio-wise difference in earnings quality ratios of banks after controlling for possible variations due to trend over time While taking each ratio separately among banks, it is seen from the multivariate tests that the difference in all ratios is highly significant as the P values (0.000) in all the indicators are less than c. Degree of variation in ratios While looking at the associated Partial Eta Squared, the importance of the differences in the earnings quality performance among banks is seen in the following order. The value of associated Partial Eta Squared is given in brackets.

7 1) II to WF (0.988) 2) NIM (0.985) 3) OE to TI (0.969) 4) OP to WF (0.967) 5) NP to AA (0.931) 6) B to II (0.803) 7) NII to TI (0.758) 8) PGINNP (0.347) Bank sector-wise analysis Bank sector-wise analysis clearly shows that the performance of NGPB is, compared to their rivals. On the profitability side, i.e., in all the ratios, the new generation private sector banks come first, whereas on the capacity of the bank to raise interest income i.e., in the ratio of interest income to working fund, public sector banks show outstanding performance. Again on operational effectiveness, OP to WF and B to II and non-interest income to total income, NGPB again come first and in the ratio of OE to TI, public sector banks stand first Bank sector-wise analysis- Earnings quality ratios and variations among PSB and NGPB due to trend over time Hypothesis III H0: There is no significant difference in the means of earnings quality ratios of PSB and NGPB due to trend over the period of study. a. Difference in Earnings quality ratios between PSB and NGPB due to time Table 7 Difference in Earnings quality ratios between PSB and NGPB due to time Effect Value F Hypothesis Error Sig. Partial Eta Squared Time Pillai's Trace a Wilks' Lambda a Hotelling's Trace a Roy's Largest Root a a. Exact statistic b. The statistic is an upper bound on F that yields a lower bound on the significance level b. Design: GBC + Time The difference in the earnings quality performance of public sector banks and new generation private sector banks, due to time was analysed, and the results of the MANCOVA (Table 7) show that the differences in earnings quality ratios over time are highly significant at 5% level of significance as Pillai s Trace F Value is and significance value is which is less than The associated Partial Eta Squared is As the significance value is less than 0.05, the null hypothesis is rejected and so the difference is highly significant. b. Ratio-wise difference in earnings quality ratios between PSB and NGPB due to time Table 8 Variations in the means of Earnings quality ratios of PSB and NGPB due to time Sl. No. Ratio F value Significance value Partial Eta Squared 1 OP2WF PGINNP NIM NP to AA II to WF OE to TI NII to TI B to II While individually considering the earnings quality ratios of PSB and NGPB, it is seen that the variation is significant in the ratios namely NP to AA, II 2 WF, NII to TI, and burden to interest IRJBM ( ) Volume No VII September Issue 9 Page 24

8 income, as the P value is less than The difference is found insignificant in the other four ratios, namely, OP to WF, PGINNP, NIM, OE to TI, between the two categories of banks. Among them, II to WF seems to vary more among PSB and NGPB as the Partial Eta Squared is which is higher than in all other ratios. c. Degree of variation in ratios Further, while looking at the value of associated Partial Eta Squared the importance of the difference in ratios over time is seen in the order of II to WF, B to II, NP to AA, PGINNP, NII to TI, NIM,OE to TI and OP to WF Difference in Earnings quality ratios between PSB and NGPB after controlling for possible changes due to trend over time a. Difference in earnings quality performance after controlling time Hypothesis IV H0: There is no significant difference in the means of earnings quality ratios of PSB and NGPB after controlling for possible changes due to trend over the period of study. Table 9 Earnings quality ratios -Difference between PSB and NGPB after controlling for time Hypothesis Partial Eta Effect Value F Error Sig. Squared GBC Pillai's Trace Wilk s Lambda a Hotelling's Trace Roy's Largest Root b a. Exact statistic b. The statistic is an upper bound on F that yields a lower bound on the significance level c. Design: GBC + Time The result of the Multivariate analysis of co-variance (MANCOVA, Table 9) reveals that Pillai s Trace is which is significant as the P value is which is less than 0.05 and the associated Partial Eta Squared is Therefore the difference in the earnings quality performance of PSB and NGPB after controlling for possible changes due to trend over time is highly significant. Hence the null hypothesis is rejected. b. Ratio-wise difference in Earnings quality ratios between PSB and NGPB after controlling for possible changes due to trend over time While taking each ratio separately among banks, it is seen from the Multivariate tests that the difference in all ratios is highly significant as P values (0.000) in all the indicators are less than c. Degree of variation in ratios While looking at the associated Partial Eta Squared, the importance in terms of the differences in the earnings quality performance among banks is seen in the following order. The value of associated Partial Eta Squared is given in brackets. 1) II to WF (0.986) 2) OE to TI (0.960) 3) OP2WF (0.953) 4) NIM (0.951) 5) NP to AA (0.920) 6) B to II (0.728) 7) NII2TI (0.635) 8) PGINNP (0.325) IRJBM ( ) Volume No VII September Issue 9 Page 25

9 d. Earnings Quality Ratios and Banks Inter-Sector Comparison Hypothesis V H0: There is no significant difference in the earnings quality ratios between PSB and NGPB over the period of study Ratio -wise difference between PSB and NGPB Table 10 Earnings quality ratios Ratio-wise difference between PSB and NGPB Category of Banks Simple Contrast PSB Vs NGPB Dependent Variable OP2WF PGINNP NIM NP2AA II2WF OE2TI NII2TI B2II Contrast Estimate Hypothesized Value Difference (Estimate- Hypothesized) Std. Error Sig % Confidence Interval for Difference a. Reference category = 2 Lower Bound Upper Bound Regarding differences among public sector banks and new generation private sector banks, it is obvious from Table 10 that there is significant variation in the earnings quality ratios of OP to WF, NP to AA, OE to TI, NII to TI and burden to interest income, as the P value of all ratios is less than 0.05 at 5% significance level. There is no significant difference in the ratios of PGINNP, II to WF and NIM as the P value is more than e. Overall difference in Earnings quality ratios of PSB and NGPB Table 11 Difference in Earnings quality ratios between PSB and NGPB Effects Value F Hypothesis Error Df Sig. Partial Eta Squared Pillai's trace a Wilks' lambda a Hotelling's trace a Roy's largest root a a. Exact statistic IRJBM ( ) Volume No VII September Issue 9 Page 26

10 Further looking at the MANCOVA results (Table 11), we may say that the variations in the earnings quality ratios of public sector banks and new generation banks differ from each other as the concerned Pillais Trace F value is which is highly significant at 5 % significance level as P value which is less than 0.05 and the related Partial Eta Squared is Hence the null hypothesis is rejected. CONCLUSION The post-liberalized era has witnessed for more positive changes in the banking industry in India. The banks have entered in to different areas and they can earn income from other sources than their core income. In order to survive in the market, banks should improve their profitability and this can be done only through higher operational efficiency. The study revealed that HDFC Bank Ltd is the best bank in terms of overall earnings quality performance followed by State Bank of Travancore (SBT). New generation private sector banks performance is than public sector banks. There is significant difference between the means of earnings quality ratios of selected banks, and the two sectors of banks, PSB and NGPB due to time and after controlling differences due to time. Considering the difference in individual ratios between the two sectors, no difference is found in Net Interest margin. SUGGESTIONS 1. Public sector banks should take steps to improve the growth in net profit, non interest income and return on assets. 2. New generation private banks rely more on the fee income than the core income. These banks should also concentrate more increasing the interest income. SCOPE FOR FURTHER RESEARCH 1. Impact of global financial crisis on the profitability of commercial banks in India 2. Comparison between the earnings quality performance of all bank groups 3. Interest and non-interest income of commercial banks : An analysis REFERENCES 1. Bodla B S, Richa Verma Evaluating Performance of Banks through Camel Model: A Case Study of SBI and ICICI Bank Ltd. Management, Volume (Year): V (2006) Issue (Month): 3 (August) Pages: Indian Banking at a Glance, The Analyst Special Issue, Vol.XVI Issue 10, October 2010 IUP Publications, ISSN Hyderabad, Pp Kothari C.R, Research Methodology, New Age International publishers New Delhi 2 nd Revised edition Dr.Nageswar Rao and Dr.Shefali Tiwari, Efficiency Indicators of Commercial banks in the Liberalised Environment in India: Abhigyan Vol.XXVII No.1 pp Ram Mohan, T.T., Ray, S. (2004), Productivity growth and efficiency in Indian banking: A comparison of public, private and foreign banks, Working Paper No. 2004/27Department of Economics, University of Connecticut, Storrs, CT, 6. Report on trends and progress of banking in India , Annual publication RBI Mumbai 7. Statistical tables relating to banking in India various years, RBI Mumbai. 8. Surya Chandra Rao D, Banking Reforms in India An Evaluative study of Performance of Commercial Banks, Regal Publications, New Delhi 008 ISBN IRJBM ( ) Volume No VII September Issue 9 Page 27