Are bank charges a threat or opportunity to banks' intermediation function? The case of South Africa.
The conventional theory of financial intermediation maintains that it is the primary function of banks to collect deposits from surplus units--savers, typically households and further channel those funds to deficit units--borrowers, typically the enterprise sector and government (Gurley & Shaw, 1960; Schmidt, Hackethal, & Tyrell, 1997). As such, banking institutions exist as intermediaries and transformers of funds to increase the social value of capital by channeling it towards more efficient uses. In the realm of financial intermediation, Diamond (1984) postulates that banks play a special role of providing liquidity and financing investment projects of borrowers which capital markets would not be able to do efficiently. It is in this respect that banks are therefore regarded as the primary conduit between savers and borrowers for intermediation purposes. Throughout the process of financial intermediation, bank charges play an important part in defining efficient financial intermediation in the financial system.
Moreover, in light of fast growing greater competition due to fast changing market conditions, it has become imperative for banks to obtain information concerning their patronage factors in relation to the way bank charges affect their role as financial intermediaries. According to Gurley and Shaw (1960) and Hestre (1969), financial intermediation is a process where financial transactions between borrowers and savers take place through the banking system. Even under conditions where bank charges are high, the banks' role as financial intermediaries should remain efficient in order to enhance well-functioning financial system in the economy.
In context of the South African economy, the banking sector system is highly concentrated and sophisticated with the five largest banks; namely Absa, First National Bank, Investec, Nedbank and Standard Bank. In absolute terms, these five banks account for between 70 percent and 90 percent of the market share of the banking industry's assets (Ojah, 2005). The country's banking sector consists of a high concentration of corporate ownership with most of the large insurance and other non-bank financial institutions (NBFIs) being either controlled by banks or the NBFIs themselves having controlling interest in banks. This reveals both complexity and intensity of competition in the industry, which should therefore enhance efficient financial intermediation.
Banks therefore need to be efficiently fully engaged in financial intermediation in a manner that embraces customers and at the same time avoiding emergence of disintermediation. This helps to ensure that banks remain serving as channels through which financial system consolidation realizes its full potential in curbing financial instability, and thus increasing the economic welfare of the economy's citizens (Ojah, 2005).
As part of the key focus area of this study, this research investigates whether bank charges, based on customers' perceptions, are either an opportunity or threat to their intermediation role in the context of the South African economy. The rationale behind the study is to examine whether the respective banking industry is currently experiencing a general trend towards financial disintermediation due to high bank charges as well as increasing options for investing funds in alternative investments in other markets. Additionally, the study extends to examine whether it is substantially realistic that banks in South Africa seem to be discouraging a savings culture given the prevailing levels of bank charges or transaction fees charged to deposit and withdraw funds over customer accounts.
In light of the above, the fundamental question to be answered is whether bank charges are either an opportunity or threat to their role as financial intermediaries. From this principal question follows the persistent disaggregated questions of the possible economic implications of bank charges for the banking industry and efficient leveraging of financial intermediation. It is against that background that the study therefore becomes imperative to understand whether bank charges are providing as an opportunity or threat to their role as financial intermediaries.
The literature on bank charges and their role on financial intermediation or disintermediation in emerging economies remain subjective. According to Allen and Parwada (2004), banks key roles are to mobilize financial resources, lend funds and facilitate payment services of financial transactions.
They play a wide number of roles which include facilitation payments between customers by providing payment systems, transmission of monetary policy by acting as conduits through which central bank's monetary policy impact the financial sector, credit allocation to particular units of the economy, making investments more accessible to individual investors and issuing of financial claims (Saunders & Cornett, 2008).
In the realization that the primary function of banks is to mobilize savings from surplus units and allocate these funds among competing users and deficit units on the basis of expected return and risk trade-offs, they therefore act as catalysts for economic growth (Pati & Shome, 2006). Thus failure by banks to efficiently execute this crucial intermediation role leads to far reaching repercussions on economic development as experienced by the global financial crisis of 2007- 2008. Levine (2002) concurs with the view that financing, both bank-based and market-based, is essential for economic growth and that financial development enhances efficiency in the allocation of scarcely available productive resources, thus stimulating the growth process.
In the study by Rajan and Zingales (1998), it was examined whether a link exists between financial development and economic growth, specifically investigating whether financial development facilitated economic growth by reducing the costs of transactions and external financing. The results from the study indicate that financial development influences economic growth through reduction of transaction and external financing costs.
Under circumstances where intermediation is not efficiently achieved, the end-result becomes financial disintermediation. Financial disintermediation is a movement away from the intermediated provision of financial services via banks to direct financial relations between lenders and borrowers. This reduces and eventually leads to the collapse of financial intermediaries in a financial system of an economy. In the South African economy, the financial sector is comprised of capital markets, the banking sector and the NBFIs (Mpako, 2007).
The country has an established, well regulated and sophisticated financial sector made up of banking; insurance and securities industries that are highly interdependent. Financial service providers comprise of banks, insurance companies, pension funds, unit trusts, fund managers, underwriters and investment banks (Edwards & Mishikin, 1995).
In view of the above, a preceding research investigation was also conducted by Mpako (2007) on whether the significance of South African banks as financial intermediaries was declining based on growth and substitutability of bank deposits by money market mutual funds. Money market unit trusts are specialized money market intermediaries that purchase money market instruments by pooling funds from investors to enable investors to earn high favorable yields from money market instruments comparative to deposits at banks. From the results, Mpako (2007) concluded that bank deposits as a ratio of total bank assets had been declining over the period 1997 to 2007, but mainly due to currency crisis as opposed to bank charges.
However, the research did not empirically validate whether disintermediation was taking place in the South African banking sector; although the regression relationships shown had provided as a strong tool for analyzing trends of bank deposits. The study focused specifically much on bank deposits and thus did not test whether the traditional intermediation role of banks was declining due to high alleged bank charges.
Another preceding study on testing for bank disintermediation by Kaufman & Mote (1994) was conducted and indicated that the role of South African banks as financial intermediaries was declining from an asset side perspective; coupled with a remarkable rise in the assets of NBFIs especially in terms of their share of bank deposits. The study concluded that disintermediation was taking place from the liabilities or bank deposit perspective; suggesting that banks were not actively sourcing deposits from surplus and facilitating transactions to deficit agents in an efficient manner. The study; however did not validate the extent to which such disintermediation was taking place.
From the broader financial perspective, it is imperative to consistently investigate other primary factors that influence banks' role as financial intermediaries in the economy other than concentrating on the traditional approach of analysing trends of bank deposits (King & Levine, 1993). The banking services industry is characterised by high levels of credence and experiential features that make them difficult to be evaluated before consumption. Therefore, in order to minimize the risk and uncertainty related to the purchase of the banking service, customers consider a range of options that yield the highest possible return for their funds in form of bank charges, alternative investments and costs of switching from one bank or account to another in search lower costs and high returns (Gurley & Shaw, 1960).
According to Allen and Santomero (2001), both bank charges and the costs associated with switching from one bank to another are principal factors influencing banks intermediation role. In that sense, a switching barrier is therefore any factor that makes it difficult or costly for customers to change providers of banking and financial intermediation services. A switching cost can also be regarded as the technical, financial or psychological factor which makes it difficult or expensive for a customer to change brand. When such costs are high for the customer, there is a greater probability that the customer will remain loyal in terms of repeated banking behavior because of the risk involved in switching between bank accounts (Metron & Bodie, 1995).
A 5-point interval Likert scale was used to measure the respondents' perception on whether there is an association between bank charges and banks intermediation function. The Likert scale examines how strongly respondents agree (5) (Strongly agree) or disagree (1) (Strongly disagree) with statements that measure variables in the hypotheses of the study. Finally, reliability and validity tests of the measures from the questionnaire are conducted to obtain reliable data that can be analysed properly to reveal meaningful findings.
Sample and Data Collection
A structured questionnaire was used to collect primary data on perceptions of respondents about the association of bank charges with banks intermediary role. The selected respondents represent a reasonable sample size with a balanced mix of demographic factors; namely age, gender, education levels, employment status and income levels. A sample of 200 respondents was used to collect data for the analysis. The questionnaire was self-administered by the researcher.
The chi-square distribution test technique was applied to test for association between banks intermediation role and a range of respondents' perceptions on bank charges. The Statistical Package for Social Science (version 19) is used for the analysis. The raw data are first factor analysed to summarise the sixteen variables into smaller sets of linear composites that preserve most of the information in the original data set. The data are subjected to Principal Component Analysis to reduce components through varimax rotation. The factors to be used in the analysis are to be constructed based on Factor Analysis (available from author upon request). Reliability test based on Cronbach's alpha is conducted from a total of sixteen factors. From the total of sixteen factors, seven are constructed and taken for analysis; described as follows.
1) Relationship- Level of bank charges has adversely affected my relationship with my bank
2) Investment consideration--I am considering moving my money to other forms of investments because my money has been eroded away by bank charges
3) Mattress option--Given an option, I would keep my money under the mattress if bank charges remain too high
4) Other options of keeping money--The high levels of bank charges have influenced me to consider other options of keeping my money
5) Choice--The levels of bank charges have a bearing on the bank I choose to bank with
6) Charges determine future choice--In future, I will consider the level of bank charges before I choose a bank to bank with
7) Moving money--I will consider moving money from my current bank to another if I perceive my bank charges to be high
RESULTS AND ANALYSIS
From the descriptive analysis results presented (Table 1) below, the majority of respondents strongly agreed that the levels of bank charges have negatively affected their relationships with their banks and given an option, they would keep their money under mattress if bank charges remain high.
The majority of respondents also regarded bank charges as not an important determining their choices of the banks they can choose to bank with. Furthermore, most respondents strongly disagree that they can consider moving their money to other forms of investments purely because their money has been eroded away by bank charges.
With respect to distribution, results above indicate that most observations/values of investment consideration, other options of keeping money, charges determine future choice and choice are concentrated on the left side of the mean, with extreme values to the right; hence their distributions are right skewed. This is indicated by the skewness statistics of the respective constructs which are all greater than zero. On the other hand, most values of relationship, mattress option and moving money are concentrated on the right of the mean, with extreme values to the left; hence the distributions are left skewed. This is shown by skewness statistics that are all less than zero.
The raw data collected was first analysed using factor analysis technique with SPSS 19.0 to summarize the 16 variables into smaller sets of linear composites that preserved most of the information in the original data set. The data was subjected to principal component analysis, a method categorized under the broad area of factor analysis. The 16 variables were reduced to seven principal components through varimax rotation (Table 2). The seven factors emerged with no cross-construct loadings above 0.5; indicating good discriminant condition.
The data also demonstrated convergent validity with factor loadings exceeding 0.5 for each construct. Consequently, the results confirm that each of the selected seven constructs is uni-dimensional and factorially distinct; hence all items used to operationalize a particular construct are loaded onto a single factor.
From the results of the analysis, seven factors with eigenvalues greater than 1.0 were obtained and they accounted for 62.23% of the total variance. In the study, a pre-analysis testing for suitability of the entire sample for factor analysis was also applied.
As shown in Table 3, the Kaiser- Meyer-Olkin (KMO) measure of sampling adequacy is 0.478 and the Bartlett's test of sphericity 217.711, significant at p < 0.000. This indicates that the sample used in the analysis is suitable for factor analytic procedures. Moreover, as the chi-square test statistic is 217.112 (p-value = 0.000), the study identified that there was extremely low probability of obtaining this result (a value greater than or equal to the obtained value) if the null hypothesis was true. The study assumed the null hypothesis as the population correlation matrix of the measures is an identity matrix. Hence, the null hypothesis is rejected as the variables were correlated with each other.
The chi-square non-parametric test of statistical significance for bivariate tabular analysis was applied to examine whether there is association between banks intermediation role and each of the variable explored in factor analysis above. The principal rationale and advantage associated with this technique is that it provides the degree of confidence to either accept or reject the null hypothesis of association between given variables.
Typically, the hypothesis tested with the chi-square technique is whether or not two different samples are different enough in some characteristic or aspect of their behaviour that can be generalized from the given samples that the populations from which the samples are drawn are also different in the behaviour or characteristics.
From the results indicated in Table 4, the hypothesis of association between differences in each of the variables defined and intermediation role is rejected for all constructs. The results: [chi-square.sub.(0.05;4)] = 3.831, p = 0.429 for Relationship; [chi-square.sub.(0.05;4)] = 2.481, p = 0.648 for Choice; [chi-square.sub.(0.05;4)] = 6.596, p = 0.159 for Mattress option; chi-square(a05; 1) = 0.733, p = 0.392 for Money moving; chi-square(a05; 3) = 2.799, p = 0.424 for Investment consideration; [chi-square.sub.(0.05;4)] = 0.684, p = 0.953 for charges determine future choice; and [chi-square.sub.(0.05;4)] = 2.698, p = 0.610 for other options of keeping money, all suggest that there is no statistically significant relationship between banks intermediation role and the constructs identified (customers perceptions on bank charges).
This implies that even if the level of bank charges has adversely affected customers' relationships with their respective banks and would therefore consider moving their money to other forms of investments and also keep their money under mattress due to erosion of the purchasing power of their funds, that does not imply that banks role of financial intermediation is diminishing. The empirical results therefore get us to the conclusion that bank charges are not a threat to banks intermediation role.
The Cramer's V symmetrical measure technique has been applied to measure the extent of association or relationship between banks intermediation role and the variables used in the study analysis. The Cramer's has been applied for its fitness as a post-test that determine strengths of association after chi-square has determined significance association between given variables. Cramer's V varies between 0 and 1; a value close to 0 shows weak association between variables while values close to 1 indicate strong association between given constructs.
Results presented in Table 5 indicate very weak relationships between the variables used in the analysis and banks intermediation role. The Cramer's values results: V = 0.170 for Relationship; V = 0.137 for Choice; V = 0.223 for Mattress option; V = 0.074 for Money moving; V = 0.145 for Investment consideration; V = 0.072 for Charges determine future choice and V = 0.142 for Other options of keeping money, all suggest very weak relationships between banks intermediation role and the constructs identified (customers perceptions on bank charges). Thus, the Cramer's V results indicate very weak strength between banks intermediation role and the constructs identified support chi-square results of no association in differences between the constructs under analysis.
RECOMMENDATION FOR FUTURE RESEARCH
In the further study around the same area, a different approach and methodology will be applied to sufficiently explore and measure primary factors influencing banks financial intermediation role from a macroeconomic perspective. Non-structural methods of measuring concentration and competition levels in the banking sector will also be used to yield more insights on the relationship between banks conduct and their intermediation role.
Empirical results from this study did not find substantial statistical significance of any association between bank charges and banks financial intermediation role. In the analysis rather, it was found that factors such as customers relationships with their respective banks, alternative investments options, money keeping decisions and choice of bank; all being influenced by the level of bank charges, do not have a bearing on banks intermediation role in respect of customers' perceptions. This further suggests that if the banking sector is alleged to be experiencing diminishing trend in its role of financial intermediation, then it could significantly be due to some factors other than either bank charges themselves or customers' perceptions on bank charges. It is therefore against this background safe to conclude that banks in South Africa stand a great chance of advancing and playing an effective financial intermediation role that can contribute remarkably to economic growth and development, thus reducing unemployment and poverty levels in the economy.
APPENDIX I Communalities Initial Extraction Relationship 1.000 .648 Feeling 1.000 .447 Choice 1.000 .747 Moving money 1.000 .607 Mattress option 1.000 .633 Investments 1.000 .621 Investments consideration 1.000 .730 Institutions 1.000 .631 Banning bank charges 1.000 .276 Bank services outside 1.000 .781 South Africa Prefer cash basis 1.000 .543 Other options of keeping 1.000 .691 money Charges determine future 1.000 .764 choice Intermediation role 1.000 .693 diminishing Should not charge us 1.000 .609 Banks play an important 1.000 .536 role hence I do not mind them charging me Extraction Method: Principal Component Analysis. APPENDIX II Component Matrix (a) Component 1 2 3 4 5 Relationship -.475 .557 -.182 -.009 -.148 Feeling .153 -.197 -.468 .339 .017 Choice -.346 .430 .001 .045 .198 Moving money -.278 -.045 .183 .321 .132 Mattress option -.231 .288 -.576 .074 .266 Investments .034 .496 -.132 -.457 -.267 Investments consideration .718 .242 -.081 -.125 -.219 Institutions .444 .270 .305 .340 .289 Banning bank charges .307 .272 -.127 -.151 .093 Bank services outside .661 .169 .068 .124 -.446 South Africa Prefer cash basis .462 .035 -.103 .352 .401 Other options of keeping -.100 .268 .210 .487 -.086 money Charges determine future .135 -.057 .270 -.431 .565 choice Intermediation role -.239 -.308 .379 -.035 -.366 diminishing Should not charge us .079 .230 .373 -.344 .275 Banks play an important -.183 .452 .448 .291 -.104 role hence I do not mind them charging me Component 6 7 Relationship .045 -.234 Feeling .067 -.215 Choice .601 .200 Moving money .338 -.509 Mattress option .121 .272 Investments -.266 .076 Investments consideration .288 .042 Institutions -.258 -.042 Banning bank charges .128 -.208 Bank services outside .309 .014 South Africa Prefer cash basis .160 .090 Other options of keeping -.389 .411 money Charges determine future .169 .370 choice Intermediation role .451 .245 diminishing Should not charge us -.077 -.459 Banks play an important .042 .016 role hence I do not mind them charging me Extraction Method: Principal Component Analysis. (a.) 7 components extracted.
Allen, D.E., & Parwada, J.T. (2004). Effects of banks fund management activities on the disintermediation of bank deposits. Journal of Business Finance and Accounting, 31(7), 1151-1170.
Allen, F., & Santomero, A.M. (2001). What do financial intermediaries do? Journal of Banking and Finance, 25(2), 200-217.
Diamond, D.W. (1984). Financial intermediation and delegated monitoring. The Review of Economic Studies, 51(3), 393-414.
Edwards, F.R., & Mishikin, F.S. (1995). The decline of traditional banking Implications for financial stability and regulatory policy, Federal Reserve Bank of New York.
Gurley, J.G., & Shaw, E.S. (1960). Financial intermediaries and the saving-investment process. The Journal of Finance, 11(2), 257-276.
Hestre, D.D. (1969). Financial disintermediation and policy. Journal of Money, Credit and Banking, Conference of University Professors, 1(3), 600-617.
Kaufman, G.G., & Mote, L.R. (1994). Is banking a declining industry? A historical perspective. Economic Perspectives, 18(3), 2-21.
King, R.G.,& Levine, R. (1993). Finance and growth: Schumpeter might be right. The Quarterly Journal of Economics, 19(9), 971-93.
Levine, R. (2002). Bank-based or market-based financial systems: which is better? The Journal of Financial Intermediation, 11(2), 398-428.
Merton, R., & Bodie, Z. (1995).A functional perspective of financial intermediation, Journal of Financial Management, 24(2), 23-41.
Mpako, V.M.W. (2007). Money market mutual funds and their impact on bank deposits in South Africa, University of Pretoria.
Ojah, K., & PILLAY, K. (2009). Debt markets and corporate debt structure in an emerging market: The South African example. Economic Modeling, 26(6).
Pati, A.P., & Shome, D. (2006).Disintermediation in financial services: an empirical investigation of the Indian banking sector. The Indian Journal of Commerce, 59(3).
Rajan, R.G., & Zingales, L. (1998). Financial dependence and growth, The American Economic Review, 88(3), 559-586.
Saunders, A., & Cornett, M.M. (2008). Financial institutions management: A risk management approach (6th Edition), 9-10.
Schmidt, R.H., Hackethal, A., & Tyrell, M. (1997). Disintermediation and the role of banks in Europe. Journal of Financial Intermediation, 2,36-67.
University of South Africa
Charles Nyoka a current Financial Doctoral student with the University of South Africa is a Lecturer at the same university and is currently lecturing Banking modules in the Department of Finance, Risk Management and Banking. He completed his MBA in Finance and Strategy at the University of Zimbabwe in 1995. He became an Associate member of the Institute of Bankers Zimbabwe the same year and was the Chairperson of the Financial Traders Association of Zimbabwe for two, three year consecutive terms. Charles has a passion for training in Corporate Finance, Strategic Management, Financial Accounting, Relationship Marketing, Risk Management, and Strategic Cash Management and Team building.
Table 1 Descriptive Statistics Relationship Investments Mattress consideration option N Valid 133 133 133 Mean 4.45 2.60 3.51 Std. Error of Mean .079 .056 .124 Mode 5 3 5 Std. Deviation .917 .651 1.428 Variance .840 .423 2.040 Skewness -1.681 .118 -.467 Std. Error of Skewnes .210 .210 .210 Kurtosis 2.058 -.277 -1.183 Std. Error of Kurtosi .417 .417 .417 Other Charges Choice Moving options of determine money keeping future money choice N Valid 133 133 133 133 Mean 3.24 2.88 3.06 .91 Std. Error of Mean .091 .083 .089 .025 Mode 3 2 3 1 Std. Deviation 1.046 .954 1.028 .288 Variance 1.093 .910 1.057 .083 Skewness .390 .670 .048 -2.893 Std. Error of Skewnes .210 .210 .210 .210 Kurtosis -.856 -.273 -.406 6.468 Std. Error of Kurtosi .417 .417 .417 .417 Table 2 Factor Analysis--Total Variance Explained Component Initial Eigenvalues Total % of Cumulative Variance % 1 2.083 13.021 13.021 2 1.528 9.551 22.571 3 1.365 8.532 31.103 4 1.354 8.462 39.565 5 1.268 7.926 47.492 6 1.246 7.785 55.277 7 1.112 6.950 62.228 8 .987 6.168 68.395 9 .906 5.662 74.057 10 .853 5.333 79.390 11 .745 4.656 84.046 12 .679 4.242 88.288 13 .558 3.489 91.777 14 .531 3.319 95.096 15 .455 2.844 97.940 16 .330 2.060 100.000 Component Extraction Sums of Squared Loadings Total % of Cumulative Variance % 1 2.083 13.021 13.021 2 1.528 9.551 22.571 3 1.365 8.532 31.103 4 1.354 8.462 39.565 5 1.268 7.926 47.492 6 1.246 7.785 55.277 7 1.112 6.950 62.228 8 9 10 11 12 13 14 15 16 Component Rotation Sums of Squared Loadings Total % of Cumulative Variance % 1 1.795 11.222 11.222 2 1.510 9.437 20.659 3 1.402 8.762 29.421 4 1.361 8.504 37.924 5 1.333 8.334 46.258 6 1.296 8.099 54.358 7 1.259 7.870 62.228 8 9 10 11 12 13 14 15 16 Extraction Method: Principal Component Analysis Table 3 KMO and Bartlett's Test Kaiser-Meyer-Olkin Measure .478 of Sampling Adequacy. Bartlett's Test Approx. Chi-Square 217.711 of Sphericity d.f 120 Sig. .000 Table 4 Chi-Square Test Results: Intermediation Role--(Dependent Variable) Variable (construct) Pearson's d.f chi-square value Relationship vs. Intermediation role 3.831 4 Choice vs. Intermediation role 2.481 4 Mattress option vs. Intermediation role 6.596 4 Moving money vs. Intermediation role 0.733 1 Investment consideration vs. 2.799 3 Intermediation role Charges determine future choice vs. 0.684 4 Intermediation role Other options of keeping money vs. 2.698 4 Intermediation role Variable (construct) Sig. Decision Relationship vs. Intermediation role 0.429 Reject [H.sub.0] Choice vs. Intermediation role 0.648 Reject [H.sub.0] Mattress option vs. Intermediation role 0.159 Reject [H.sub.0] Moving money vs. Intermediation role 0.392 Reject [H.sub.0] Investment consideration vs. 0.424 Reject [H.sub.0] Intermediation role Charges determine future choice vs. 0.953 Reject [H.sub.0] Intermediation role Other options of keeping money vs. 0.610 Reject [H.sub.0] Intermediation role Table 5 Cramer's V Results Variable (construct) Cramer's Sig. Decision V score Relationship vs. Intermediation 0.170 0.429 Very weak role Choice vs. Intermediation role 0.137 0.648 Very weak Mattress option vs. Intermediation 0.223 0.159 Very weak role Moving money vs. Intermediation 0.074 0.392 Very weak role Investment consideration vs. 0.145 0.424 Very weak Intermediation role Charges determine future choice 0.072 0.953 Very weak vs. Intermediation role Other options of keeping money vs. 0.142 0.610 Very weak Intermediation role
|Printer friendly Cite/link Email Feedback|
|Publication:||International Journal of Business, Accounting and Finance (IJBAF)|
|Date:||Sep 22, 2013|
|Previous Article:||The effects of the Securities Litigation Uniform Standards Act (SLUSA) on earnings forecasts.|
|Next Article:||Voluntary accounting disclosures in the airline industry.|