The trade off between interest income and non-interest income of Vietnam commercial banks

Some empirical researches find fee-based income activities have negative impacts to the trade-off between returns and risks while other studies get opposite results.. Williams and Prathe

The trade off between interest income

and non-interest income

of Vietnam commercial banks

VU THI LE GIANG

University of Economics HCMC – giangvtl@ueh.edu.vn

HOANG HAI YEN

University of Economics HCMC – yenhh@ueh.edu.vn

1 Introduction

The main traditional activities of banks include deposit taking and lending Besides these activities, banks also diversify their activities to non-interest income activities and this trend become more and more popular Research of Stiroh (2004) on American banks stated that in 1980s non-interest income activities account for 19% total banks’ income while 2001 this number was 43% Studies of Lepetit, Nys, Rous, and Tarazi (2008) on European banks also got the same results, the percentage of non-interest income increased from 26% to 41% This trend is also repeated in other countries such as Australia, China although the growth rate is not as high as in US and Europe The strong increasing in non-interest income raises a question for bank managers and researchers: whether increasing in non-interest income activities is good for banks? Should banks continue this trend? Many researches related to this field are conducted but results are mix findings

Vietnam also follows this trend With difficulties in credit expansion, banks are increasingly seeking revenues from non-interest income activities Results of this shift are not fully assessed Prior researches in Vietnam only studied about the sole impact of diversify income (interest income and non-interest income) to profit or to risk but these researches haven’t assessed systematically the trade-off between risk and return of interest income activities and non-interest income activities That the reason why we study about “the trade-off between interest income and non-interest income of Vietnam commercial banks”

Non-interest income include “fiduciary income, service charges, trading revenue, and fees and other income” (Stiroh, 2004) In our research, we use “Total Non-interest Operating Income” in the balance sheet of banks as a noninterest income To exploit the problems, we use two ratios to measure the trade-off between risk and return (Vi and RAP as Williams and Prather (2010) Data from 27 Vietnam commercial banks between

2006 and 2015 is collected from Bankscope We find that diversification activities to non-interest income increases return against one unit of risk However, this return will reduce

if we continue to shift to non-interest income activities This result encourages banks to invest in non-interest income activities but they also need to control these activities to avoid over-investment

Our paper is constructed into five parts The next part will be literature review about income diversification The third part describes data and methodology The fourth part discusses the results and final part is conclusions

2 Literature review

Researches about diversification banking activities to noninterest income activities got mix findings Some research conclude that shifting toward noninterest income activities can bring back higher returns for banks while other studies prove that diversification doesn’t add benefit for banks but banks have to suffer higher risks However, all the researches agreed at the same point that traditional activities are less risk than noninterest activities

DeYoung and Roland (2001) examines on 472 US commercial banks between 1988 and 1995 They tried to answer whether, how and to what degree the shifts to noninterest income affect the volatility of bank earnings Results reveal that shifting from traditional activities to non-traditional activities was increasing and bank’s return increased when they diversify to noninterest income activities However, the volatility of bank earnings was also higher Stiroh (2004) had the same point of view with DeYoung and Roland (2001) that non-interest income was more volatile than traditional income Author studied US banks from the late 1970s to 2001 to examine the affections of non-interest income to banks profit and revenue He found that there was a reduction in volatility of bank revenue growth in 1990s but this was because of the lower volatility of net interest income “rather than the diversification benefits from increased noninterest income” (Stiroh, 2004) He concluded that noninterest income was still more volatile than net

interest income Other research conducted in Europe by Lepetit et al (2008) using data from 734 banks also found the same result Authors compared risk level of banks which diversified to untraditional activities with banks which didn’t follow this strategy Results revealed that diversifying banks suffered higher risks and these risks highly correlated with commission and fee incomes than trading activities In Vietnam, (Vo & Tran, 2015) studied 37 commercial banks from 2006 – 2013 They concluded that banks increasing fee-based activities may get lower profits and higher risks than banks pursuing traditional activities

In contrast with the above researches, the following studies confirmed that shift to noninterest income activities bring benefits to banks Smith, Staikouras, and Wood (2003) studied the variability and the correlation between interest and non-interest income for banks in European countries from 1994-1998 Sample included 200 large banks having total assets over 10 billions USD and 2455 small banks The research found that shift to non-interest income make profits in European banks stable in the research period In addition, recent study of Lee, Yang, and Chang (2014) for 967 banks of 22 Asia countries from 1995 to 2009 about the impacts of non-interest income on profits and risks found that non-interest income reduced risks but did not increase profit Especially, results became complicated when bank specialization and a country's income level were considered To saving banks, profit reduced and risks increased when they shift to non-interest income activities In high income countries, these activities increased bank risks while in middle and low income countries, non-interest income activities increased profits and decreased risks

Other researches did not examine sole impacts of shifting to fee-based income activities to returns and risks but they accessed the trade-off between returns and risks Results from these researches are also mix findings Some empirical researches find fee-based income activities have negative impacts to the trade-off between returns and risks while other studies get opposite results DeYoung and Rice (2004) proved that increasing noninterest income activities lead to higher volatilities in returns They researched 4,712 commercial banks in United State from 1989 to 2001 with 37,175 observations about the relationships of non-interest income, business strategies, market conditions, technological changes and financial performance The results stated that increasing in non-interest income made the trade-off between risks and returns poorer In addition, study of Stiroh and Rumble (2006) for Financial Holding Companies – FHCs in United

State also got the same results They researched 1800 FHCs from 1997 to 2002 and found evidences about the benefits of diversification However, these benefits were offset by increasing companies’ investments in non-interest income activities which were volatility and “not necessarily more profitable than traditional activities” Stiroh and Rumble (2006) With some typical companies, shift to fee-based income decreased risk-adjusted returns This confirmed that these FHCs must accept more risks to get non-interest incomes

Besides studies stated that non-interest income had negative impacts to the trade-off between risks and returns, some studies found opposite results Williams and Prather (2010) stated there was a positive impact to the trade-off between risks and returns when banks shift from traditional activities to fee-based income activities Williams and Prather (2010) researched on 49 commercial banks which included 4 big banks accounted for 65% total assets of commercial banks in Australia and issued most of financial products, the second group was domestic banks specialized in retail finance and the third group was foreign banks They concluded that non-traditional activities were riskier than traditional activities However, combining these two activities was benefit for shareholders in diversification their portfolio and reduced risks for banks

3 Data description and methodologies

3.1 Data description

Our research uses data of 27 commercial banks in Vietnam from 2006-2015 Foreign bank, foreign bank branches, joint-venture banks aren’t included in the sample because data of these banks aren’t updated in Bankscope As statistic of State Bank of Vietnam, there are 33 commercial banks This research excludes 6 commercial banks: Global Petro Bank, National Citizen Bank, BacA Bank, KienLongBank, HDBank and Vietbank because data of these banks aren’t updated in 5 years consecutive from 2006-2015

When studying the differences between state-owned banks and other joint stock commercial banks, we use the classification of State Bank of Vietnam However, group of state-owned banks just includes 4 banks: Bank for Foreign Trade of Vietnam, Bank for Investment and Development of Vietnam, Vietnam Bank for Agriculture and Rural Development, Vietnam Joint-Stock Commercial Bank for Industry and Trade The other state-owned banks were acquired by State Bank of Vietnam during the restructuring

banking systems Therefore, if we include the data of these banks in the data of state-owned banks, the data will be biased

3.2 Methodologies

We use two indexes to measure the trade-off between risks and returns in research of Williams and Prather (2010)

First, descriptive statistic is conducted to compare non-interest income and net interest income for all banks in the sample and in 2 groups of banks (state-owned banks and other joint stock commercial banks) Results from descriptive statistic reveal the stable of these income sources and which source is the main income source of banks After that, the percentage of each income source is calculated to know the trend of shifting to non-interest income activities in Vietnam banking system

Second, in order to understand whether diversification into non-interest income activities is benefit for banks, the authors undertake three steps

Calculate correlation of five elements: net interest income against total assets, non-interest income against total assets, net non-interest income against total equity, non-non-interest income against total equity and ROE before tax Results reveal whether the combination

of both traditional and non-traditional activities can reduce bank risks or not Besides, results also state if non-interest income activities is benefit for shareholders

Calculate Vi (Williams & Prather, 2010) This index state how much risk banks have

to tradeoff for one unit of return

бi is the annual standard deviation of returns for income source i

µi is the average annual return for the income source i

Calculate RAP (Williams & Prather, 2010) which is the return premium for each unit

of risk Comparison RAP of net interest income, RAP of non-interest income and RAP of total income will give the conclusion whether shift to non-interest income bring benefit for banks or not

Vi =

бi

µi

RAPi =

ri - rf

бi

Where:

ri : the average annual return for the income source i over the study period

rf : the average annual return for the risk-free asset over the study period

бi : the annual standard deviation of returns for the income source i for the study period

Results getting from the above three steps will be discussed and summarized to find the final results for the research

4 Results

4.1 Descriptive statistic

Table 1

Indicators related to net interest income and non-interest income

Net interest income

Non-interest income/Total assets

Net interest income /Total assets

Interest income /Total income

All banks

State owned bank

Other joint stock commercial bank

Source: Calculated from Bankscope by authors

Means of net interest income aren’t different for banks in two groups However, standard deviation of net interest income for state owned banks is smaller than that value for other joint stock commercial banks About non-interest income against total assets, mean of state-owned banks is higher than the value of joint stock commercial banks but the volatility of non-interest income against total assets for second group of banks is much higher than the first group The results from table 1 also state traditional activities are main source of income for banks with the proportion of interest income in total income

is above 90%

During the 80s, non-interest incomes in United State accounted for 19% total income while in 2001 this rate was 43% (Stiroh, 2004) The same trend also happened in Europe with increasing from 26% in 1989 to 41% in 1998 (ECB2000) In Vietnam, shift to non-interest income activities is not as strength as the above regions Statistic in research period from 2006-2015, the percentage of non-interest income only accounts for 8% of total income and this proportion has been reducing since 2010 until now Before 2011, the percentage of non-interest income was about 11% of total income but in 2011-2015, it was only 5-6% These number state that there was a strong reducing in the proportion of non-interest income We don’t investigate the reasons of this reducing However, the time starting for the down trend was the same with the time the State Bank of Vietnam started

to restructure the banking systems Therefore, we question whether the restructuring banking systems had impacts to the proportion of non-interest income From the above analysis, we divide the research period into two parts: first period is from 2006 to 2010 and the second period is from 2011-2015

Figure 1: The proportion of interest income and non-interest income over total income

Source: calculated from Bankscope data

Figure 2: The percentage of net interest income and non-interest income against total assets

Source: calculated from Bankscope data

The above figures state that during the research period from 2006 to 2015, net interest income was still the main source of income for banks and two sources of income had opposite trend Therefore, we expect that these two sources of income may have negative correlation and combination these two sources of income in the investment portfolio of banks may reduce total risk

The proportion of non- interest … 0.0%

20.0%

40.0%

60.0%

80.0%

100.0%

The proportion

of non- interest income against total income

0.00%

0.50%

1.00%

1.50%

2.00%

2.50%

3.00%

3.50%

Non-interest income against total assets

4.2 The risks and returns trade-off when banks invest in both traditional

activities and fee income activities

To determine whether investing in non-traditional activities reduces risks or not, we calculate the correlation between traditional income and non-interest income The results state that the correlation of these two sources of income for the group of all banks or for the group of join stock commercial banks are negative These results confirm that combining interest income activities and non-interest income activities can reduce total risk for banks In addition, the correlations between pre-tax ROE and non-interest income against equities for these two groups of banks are positive It means that increase non-interest income will have positive impact to pre-tax ROE

The correlation between net interest income and non-interest income of state-owned bank group is small and the signs of correlation are both negative and positive The correlation between non-interest income against total assets and net interest income against total assets is -0.05 while the correlation between non-interest income against total equity and net interest income against total assets is +0.04 Therefore, from the above information, it’s hard to conclude whether the combination of interest income activities and non-interest income activities is good or not for state-owned banks

Table 2

Correlation of elements

Net interest income/tot

al assets

Non-interest income/tot

al assets

Net interest income/tot

al equity

Non-interest income/tot

al equity

Pretax ROE

ALL BANKS

Net interest income/total

Non-interest income/total

Net interest income/total

Non-interest income/total

STATE-OWNED BANKS

Net interest income/tot

al assets

Non-interest income/tot

al assets

Net interest income/tot

al equity

Non-interest income/tot

al equity

Pretax ROE Net interest income/total

Non-interest income/total

Net interest income/total

Non-interest income/total

OTHER JOINT STOCK COMMERCIAL BANKS

Net interest income/total

Non-interest income/total

Net interest income/total

Non-interest income/total

Source: calculated from Bankscope data

To determine the risks and returns trade-off, we use two indexes presented in research

of Williams and Prather (2010) The first index Vi measures level of risks against one unit

of return The results for Vi are presented in the following table

Table 3

Vi index for indicators in each bank group

Net

interest

margin

Non-interest income/total assets

Net interest income/total assets

Non-interest income/total equity

Net interest income/total equity

Total net interest income and non-interest income/total equity ALL BANKS