Description
This paper investigates the performance of South Africa's commercial banking sector for the period 2005- 2009. Financial ratios are employed to measure the profitability, liquidity and credit quality performance of five large South African based commercial banks.
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
30
African Review of Economics and Finance, Vol. 2, No. 1, Dec 2010
©The Author(s)
Journal compilation ©2010 African Centre for Economics and Finance. Published by
Print Services, Rhodes University, P.O.Box 94, Grahamstown, South Africa
A fnancial Ratio Analysis of Commercial Bank
Performance in South Africa
Mabwe Kumbirai
2#
and Robert Webb*
Abstract
This paper investigates the performance of South Africa’s commercial banking
sector for the period 2005- 2009. Financial ratios are employed to measure
the proftability, liquidity and credit quality performance of fve large South
African based commercial banks. The study found that overall bank performance
increased considerably in the frst two years of the analysis. A signifcant change
in trend is noticed at the onset of the global fnancial crisis in 2007, reaching its
peak during 2008-2009. This resulted in falling proftability, low liquidity and
deteriorating credit quality in the South African Banking sector.
1. Introduction
Commercial banks in South Africa have undergone immense regulatory and
technological changes since the attainment of constitutional democracy in 1994.
South African banks are faced with increasing competition and rising costs as a
result of regulatory requirements, fnancial and technological innovation, entry
of large foreign banks in the retail banking environment and challenges of the
recent fnancial crisis. These changes had a dramatic effect on the performance
of the commercial banks. Most studies on bank performance in South Africa
have focused on branch performance [see Oberholzer and Van der Westhuizen
(2004); O’Donnell and Van der Westhuizen, (2002); Okeahalam (2006)]. More
recently, Cronje (2007) and Ncube (2009) studied the effciency of South
African banks using Data Envelopment Analysis (hereafter DEA), studying
the periods 1997-2007 and 2000-2005 respectively. This study evaluates bank
2 Caledonian Business School, Glasgow Caledonian University, Cowcaddens Road G4 0BA,
UK
#
Corresponding author: [email protected]
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
31
performance for the period 2005-2009 using fnancial ratio analysis (hereafter
FRA). The present study is different from earlier studies in two ways: sample
coverage and methodology. Covering South Africa’s big banks in the period both
prior to, and after the 2007 subprime meltdown highlights important changes
that have occurred in the banking industry and tease out appropriate policy for
improving bank performance. Compared to extant literature we favour FRA
because it is effective in distinguishing high performing banks from others, tends
to compensates for disparities and controls for any size effect on the fnancial
variables being studied (Samad, 2004). Additionally, fnancial ratios enable us
to identify unique bank strengths and weaknesses, which in itself inform bank
proftability, liquidity and credit quality. The rest of the paper is organised as
follows: the next section offers background information on the fnancial system
in South Africa. Section 3 outlines the past studies in bank performance. The
methodology and data used are described in section 4. Section 5 presents and
analyses the results and section 6 concludes.
2. Banking in South Africa
While ranked in the top 20 of world economies by size, the South African
economy remain relatively small accounting for less than 1 per cent of global
GDP (Baxter, 2008). Despite being small by global standards, South Africa is
the economic powerhouse of Africa, leading the continent in industrial output
and mineral production and generating a large proportion of Africa’s electricity
(Brand, 2009). South Africa’s economic performance during the frst decade
of freedom was impressive, with favourable external environment and strong
domestic demand helping raise GDP growth to 5 % on average in 2004–2007 and
lowering the unemployment rate by 5% (Ramcharan, 2009).
The consistent performance ensured by a prolonged economic expansion,
supported by prudent economic policies and improving macroeconomic
fundamentals resulted in low infation, high commodity prices and increased
investor confdence. In such favourable economic conditions, the banking sector
played an essential role in the economic growth of the country. However, the
global fnancial crisis of late 2007 sharply changed the outlook for an already
slowing economy and South Africa was not immune to the impact of the global
fnancial crisis-induced economic slowdown (Baxter, 2008). Slower economic
growth in key export markets, lower commodity prices and a slowdown in
capital fows to developing countries had a negative impact on the South African
economy. Negative growth was experienced in the fourth quarter of 2008 and
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
32
the South African economy offcially entered into a recession in the frst quarter
of 2009 (SARB, 2009). The table below shows the South African economic
indicators for the period 2005-2009. The real GDP growth has been increasing
since 2000. The growth rate was 5.3% for 2006 before falling to 5.1% in 2007
when the fnancial crisis started. However as can be seen, the growth rate then
drastically fell to 3.1% for 2008 before settling on a negative 1.8 % in 2009
refecting the above mentioned effects of the fnancial crisis on the South African
economy GDP growth. Infation, which has remained outside the 3–6 % target
band since 2007, peaked to 11.5% in 2008 as shown in table 1.
Table 1: South Africa’s Key Economic Indicators
2005 2006 2007 2008 2009
Real GDP 5.0 5.3 5.1 3.1 -1.8
CPI (annual average) 3.4 4.7 7.1 11.5 7.4
CPIX (end of period) 1 4.0 5.0 8.6 10.3 Na
Unemployment rate (percent) 26.2 25.5 22.7 21.9 24.9
Broad money (end of period) 20.5 22.5 23.6 14.8 7.9
National government budget balance
(percent of GDP)
-0.6 0.4 0.8 -0.7 -4.4
National government debt (percent of
GDP)
35.2 33.0 28.5 27.3 30.2
External current account balance
(percent of GDP)
-4.0 -6.3 -7.3 -7.4 -4.4
External debt (percent of GDP) 19.0 22.1 26.6 25.9 27.3
Gross reserves 2.9 3.1 3.7 4.6 4.5
US Dollar exchange rate (rand per u.s.
dollar
6.33 6.97 6.81 9.53 Na
Sources: South African Reserve Bank; IMF, International Financial Statistics; and
IMF Staff projections. Since January 2009, a reweighed and rebased CPI replaced the
previously used CPIX (the consumer price index excluding the interest on mortgage
loans) as the targeted measure of infation.
Despite the apparent worsening of the economic indicators, South Africa
maintained strong macro-fscal stabilization measures and had progressed on
several socio-economic fronts before the crisis
1
hence the economy’s resilience
to the global fnancial crisis. Consequently, for an open economy intricately
woven into the ‘fabric’ of the global economy, South Africa weathered the global
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
33
crisis relatively well compared to other countries in the region (Baxter, 2008).
South Africa’s 1994 transition from apartheid to constitutional democracy
has been one of the most important political achievements of its time. The safe
political and business environment resulting from the attainment of democracy
opened up opportunities for new domestic entrants to the banking system (Van
der Walt, 1998). The biggest increase in new bank registrations was in 1996,
when registrations rose from 35 to 44. However, this was followed by the biggest
shakeout during 1999 when these banks faced liquidity pressures which saw them
exiting the banking system (Mboweni, 2004). Between 1996 and 2004 more than
half of these banks disappeared, including Saambou in 2002 when, like a few
other small banks, it was denied access to lender-of last resort facilities (Gilbert
et al, 2009). Mboweni, (2004) has argued that the downward spiralling of the
small and medium banks was due to consolidation in the banking sector rather
than failure of the small and medium banks. A fact also supported by Gilbert
et al, (2009:68) who states that the change in the banking scene was as a result
of the ‘‘rationalisation of the banking system whereby organisational specifc,
functional specialisation was replaced by functional diversifcation within
multifunction institutions’’.
The socio-political changes led to structural changes in the banking sector as
well as the delivery of fnancial services. The political transformation, relaxation
of exchange controls and liberalisation of the economy resulted in South Africa
increasingly becoming an important fnancial centre (Mboweni, 2004; Murinde,
2009). Such fnancial liberalisation did not only bring greater competition for
traditional domestic banks but also improved the quality and availability of
fnancial services in South Africa. In September 2004 Barclays announced its
intention to acquire a majority shareholding in ABSA, South Africa’s second
largest bank. This event signalled the UK banking giant’s return to banking
in South Africa having been forced out by economic sanctions and political
instability in 1986. These events signifed the reintegration of the South African
domestic economy into the global environment evidenced by the relaxation of
the exchange controls and the provision of fnancial services to the poor, non-
white population that was practically excluded from the formal fnancial services
in the apartheid era (Napier, 2006).
The South African banking industry is also characterised by international
links through correspondent banking relations with off-shore banks and
international institutional investments in domestic banks. Therefore, regulation
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
34
and supervision of banking is still an issue, especially with respect to fnancial
innovations such as derivative instruments and in light of the 2007-2009 global
fnancial crisis (Murinde, 2009). The 1990 Banks Act was amended in 2008,
to align the South African banking legislative framework with the principles of
Basel II. Banking regulation is aimed at a balance between enhanced stability of
the banking sector and the costs of reduced competition, struck by successive
rounds of bank regulation (Gilbert et al, 2009). Generally, South Africa’s
approach to fnancial system stability places considerable reliance on private
and market forces to achieve fnancial system stability with Rossouw (2009)
arguing that any government intervention would be at the minimum level and
only needed to contain systemic risk.
At the heart of the South African banking system is the South African
Reserve Bank (SARB), established in 1921. As at the end of 2009 the banking
sector consisted of 12 locally controlled banks, 6 foreign controlled banks and
2 mutual banks (SARB, 2009). Further, the banking sector has 42 international
banks with authorised representative offces in South Africa. The banking
sector is highly concentrated with the largest four banks holding 86.4% of the
total industry assets (SARB, 2009; Greenberg and Simbanegavi, 2009). Most
economists would analyse the level of concentration in a banking industry using
indicators such as Herfndahl–Hirschman Index, N-frm concentration ratio and
Learner’s index of monopoly (Calem and Carlino, 1991). An H-index above 0,
18 represents a highly concentrated industry that goes some way to indicating
the presence of an oligopoly. An ‘oligopoly’ can be defned as an imperfectly
competitive market structure in which a few institutions dominate the industry.
Such a description may ft the South African banking industry whose index has
been above 0.18 since 2005 (SARB, 2009).
Although the South African banking sector has been relatively insulated
from the direct impacts of the global fnancial-sector crisis through appropriate
monitoring and supervision of the domestic banking sector, the negative
contagion effects of the crisis had a negative effect on bank balance sheets
(SARB, 2008: 2009). The aggregated balance sheet of the banking sector in
South Africa equalled R1 677 billion in 2005. The sector’s balance-sheet size
then grew to R3 177 billion 2008 (135, 4 % of GDP), followed by a decline
in asset growth during 2009, ending the year at R2 967 billion (118, 5 % of
GDP). Banking sector assets comprise mainly loans and advances, followed
by derivative fnancial instruments. Home loans and term loans represent
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
35
approximately 52% of the total assets while commercial mortgages represent
9.7%. On the liabilities side, deposits constitute a signifcant percentage of
banking-sector liabilities amounting to 79, 6 % in 2008 and 85, 4 % in 2009.
Deposits by corporate customers which constitute the largest portion of banking-
sector deposits amounted to 42, 5 % in 2009, followed by retail customers and
bank deposits, which accounted for 22, 3 % and 13, 7 %.
3. Previous research on bank performance
The measurement of bank performance particularly commercial banks
is well researched and has received increased attention over the past years
(Seiford and Zhu, 1999). There have been a large number of empirical studies
on commercial bank performance around the world (see Yeh, 1996; Webb, 2003;
Lacewell, 2003; Halkos and Salamouris, 2004; Tarawneh, 2006). However,
little has been done on bank performance in South Africa. However, with the
deteriorating health of the banking institutions and the recent surge of bank
failures as a result of the current global fnancial crisis, it is justifed that bank
performance receives increased investigation from both scholars and industry
specialists.
There are two broad approaches used to measure bank performance, the
accounting approach, which makes use of fnancial ratios and econometric
techniques. Traditionally accounting methods primarily based on the use of
fnancial ratios have been employed for assessing bank performance (Ncube,
2009). However, the limitations of this method coupled with advances in
management sciences have led to the development of alternate methods such
as non-parametric DEA and parametric Stochastic Frontier Approach (hereafter,
SFA) (Berger and Humphrey, 1997).
Berger & Humphrey (1997) assert that the whole idea of measuring bank
performance is to separate banks that are performing well from those which
are doing poorly. They further indicated that, “evaluating the performance of
fnancial institution can inform government policy by assessing the effects of
deregulation, mergers and market structure on effciency” (p175). Bank regulators
screen banks by evaluating banks’ liquidity, solvency and overall performance
to enable them to intervene when there is need and to gauge the potential for
problems (Casu et al, 2006). On a micro-level, bank performance measurement
can also help improve managerial performance by identifying best and worst
practices associated with high and low measured effciency.
This brief review will focus upon studies on South Africa and other
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
36
emerging economies. When looking to improve their performance, banks
compare the performance of their peers and evaluate the trend of their fnancial
performance over time. Tarawneh (2006) in his study measured the performance
of Oman commercial banks using fnancial ratios and ranked the banks based
on their performance. The study utilised FRA to investigate the impact of asset
management, operational effciency and bank size on the performance of Oman
commercial banks. The fndings indicated that bank performance was strongly
and positively infuenced by operational effciency, asset management and bank
size.
In the Gulf, Samad (2004) investigated the performance of seven locally-
incorporated commercial banks during the period 1994-2001. Financial ratios
were used to evaluate the credit quality, proftability, and liquidity performances.
The performance of the seven commercial banks was compared with the
banking industry in Bahrain which was considered a benchmark. The article
applied a Student’s t-test to measure the statistical signifcance for the measures
of performance. The results revealed that commercial banks in Bahrain were
relatively less proftable, less liquid and were exposed to higher credit risk than
the banking industry, in which wholesale banks are the main component.
Kiyota (2009) in a two- stage procedure investigated the proft effciency
and cost effciency of commercial banks operating in 29 Sub-Saharan African
countries during 2000-2007. The article employs the SFA for the estimation of
proft and cost effciency, fnancial ratios and the Tobit regression to provide
cross-country evidence on the performance and effciency of African commercial
banks. The fndings based on a range of performance ratios as well as stochastic
cost and proft frontier estimation, suggest that foreign banks tend to outperform
domestic banks in terms of proft effciency as well as cost effciency. The results
are also in line with the research by Kirkpatrick et al (2007) who used a sample of
89 banks from Sub-Sahara African countries for the period 1992-1999 and found
that banks are on average 67% proft effcient and 80% cost effcient, as indicated
by the results from both the distribution free approach and SFA methods.
O’Donnell and Van der Westhuizen (2002) measured the effciency of a
South African bank at branch level. Their main focus was investigating branches
which were performing well and those that were doing badly, where effciency
could be improved. They found that many branches were operating on a scale
that is too small and could increase their operational scales thereby improving
the overall effciency of the bank. In a similar approach, Okeahalam (2006)
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
37
used the Bayesian SFA to assess the production effciency of 61 bank branches
in nine provinces of South Africa. The fndings of the study points to the fact
that although every branch is operating at increasing returns to scale, the bank
branches can reduce their cost by 17% if they improve the level of effciency.
Overall the article concludes that the bank branches are less effcient than they
should be and could obtain cost reductions by increasing output.
Further, Oberholzer and Van der Westhuizen (2004) investigated the
effciency and proftability of ten banking regional offces of one of South
Africa’s larger banks. This study demonstrates how conventional proftability
and effciency analyses can be used in conjunction with DEA. Although their
study concentrated on banking regions, their fndings confrm those of Yeh
(1996) that DEA results as an effciency measure have a relationship with both
proftability and effciency ratios. The conclusions were that there are signifcant
relationships between conventional proftability and effciency measures and
allocative, cost and scale effciency and no signifcant relationship with technical
effciency.
Most of the above mentioned studies concentrated on branches of a single
bank, one of the studies that investigated the entire South African banking
sector is Cronje (2007) who employed the DEA method and a sample of 13
South African banks to provide a measure of the effciency of the South African
banks. His fndings show that out of the 13 banks, the three largest banks are
effcient and serve as a standard for the banks classifed as ineffcient. The fourth
largest bank showed a slight ineffciency. Overall, seven banks were classifed as
ineffcient and the article recommends target areas for the banks to improve their
effciencies with guidelines that bankers in ineffcient banks could use to increase
their sustainable proftability. The results of this study are in sharp contrast to
studies in the UK where Drake (2001) and Webb (2003) found the larger banks
less effcient. This difference could be attributed to the differences in operating
environment as South Africa is an emerging economy with a different political
and economic history where as UK is a developed country.
Another study that provides a brief but interesting account of bank
performance was conducted by Ncube (2009) who uses the stochastic frontier
model to analyse the cost and proft effciency of four large and four small South
African banks. The results of the study show that South African banks have
signifcantly improved their cost effciencies between 2000 and 2005 with the
most cost effcient banks also being most proft effcient. However, effciency
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
38
gains on proftability over the same time period were found not to be signifcant.
4. Methodology and Data
This paper uses a descriptive fnancial ratio analysis to measure, describe and
analyse the performance of commercial banks in South Africa during the period
2005-2009. Additionally, to examine whether the difference in performance of
the banks in 2005-2006 is statistically different from that of 2008-2009 a student’s
t- test is employed to test the hypothesis that the means of the two periods are the
same on the seven variables as detailed in section 4.1.
The following hypothesis has been tested:
H
0:
µ
1
=µ
2
,
Where µ
1
is the mean for 2005-2006 and µ
2
is the mean for 2008-2009.
Inferences about the hypothesis are made by looking at test statistics and critical
values associated with the mean. If P-value ? ?, reject the null hypothesis. If
P-value > ?, do not reject the null hypothesis. The results of the test are to be
handled with caution as there are very few observations and the statistical tool
might not be very effective when the sample is small.
The selection of the FRA method for this study is motivated by the fact
that from the review of past studies on banking in South Africa and to the
researchers’ knowledge, no researcher has used FRA to measure the performance
of commercial banks in South Africa during 2005-2009. Authors Oberholzer and
Van der Westhuizen (2004) used the method in measuring branch performance
however, the authors concentrated on branches of a single large bank. The
main advantage of FRA is its ability and effectiveness in distinguishing high
performance banks from others and the fact that FRA compensates for disparities
and controls for any size effect on the fnancial variables being studied (Samad,
2004). Additionally, fnancial ratios can be used to identify a bank’s specifc
strengths and weaknesses as well as providing detailed information about bank
proftability, liquidity and credit quality policies (Hempel et al, 1994: Dietrich,
1996). FRA permits a historical sketch of bank returns and risks which Hempel
et al, (1994) suggests presents an opportunity to evaluate the past performance
of the bank which is an important step for planning for future performance.
Although accounting data in fnancial statements is subject to manipulation and
fnancial statements are backward looking, they are the only detailed information
available on the bank’s overall activities (Sinkey, 2002). Furthermore, they are
the only source of information for evaluating the management’s potential to
generate satisfactory returns in future.
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
39
The population for this research comprise of all the banks operating in
South Africa between 2005 and 2009. A sample of the top fve commercial
banks was selected based on the value of their total assets at the end of the
2009 fnancial year end. These are the banks that dominate the sector with the
top 4 banks controlling 84.6% of the total industry assets which makes them
systemically important banks. The fact that all banks could not be included in the
study constrains the validity of the study. However, only the fve largest South
African banks (FirstRand Bank, Absa, Nedbank, Standard Bank and Imperial
bank
2
) offer a comprehensive variety of fnancial services right through South
Africa, all the other banks are aimed at niche markets or confned to geographical
operations. The data was obtained from Bankscope and the bank’s fnancial
statements and websites.
4.1 The Variables
A. Proftability Performance
The most common measure of bank performance is proftability. Proftability is
measured using the following criteria:
Return on Assets (ROA) = net proft/total assets shows the ability of
management to acquire deposits at a reasonable cost and invest them in proftable
investments (Ahmed, 2009). This ratio indicates how much net income is
generated per £ of assets. The higher the ROA, the more the proftable the bank.
Return on Equity (ROE) = net proft/ total equity. ROE is the most important
indicator of a bank’s proftability and growth potential. It is the rate of return to
shareholders or the percentage return on each £ of equity invested in the bank.
Cost to Income Ratio (C/I) = total cost /total income measures the income
generated per £ cost. That is how expensive it is for the bank to produce a unit of
output. The lower the C/I ratio, the better the performance of the bank.
B. Liquidity performance
Liquidity indicates the ability of the bank to meet its fnancial obligations in a
timely and effective manner. Samad (2004:36) states that ‘‘liquidity is the life
and blood of a commercial bank’’. Financial liabilities are attracted through
retail and wholesale distribution channels. Retail generated funding is
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
40
considered less interest elastic and more reliable than deposits attracted
from wholesale distribution channels (Thygerson, 1995). The following ratios
are used to measure liquidity.
Liquid assets to deposit-borrowing ratio (LADST) = liquid asset/customer
deposit and short term borrowed funds. This ratio indicates the percentage of
short term obligations that could be met with the bank’s liquid assets in the case
of sudden withdrawals.
Net Loans to total asset ratio (NLTA) = Net loans/total assets NLTA measures
the percentage of assets that is tied up in loans. The higher the ratio, the less
liquid the bank is.
Net loans to deposit and borrowing (NLDST) = Net loans/total deposits and
short term borrowings. This ratio indicates the percentage of the total deposits
locked into non-liquid assets. A high fgure denotes lower liquidity.
C. Asset Credit Quality (Credit Performance)
While it is expected that banks would bear some bad loans and losses in their
lending activities, one of the key objectives of the bank is to minimize such
losses (Casu et al, 2006). Credit performance evaluates the risks associated with
the bank’s asset portfolio i.e. the quality of loans issued by the bank. Several
ratios can be used for measuring credit quality however, not all information on
the loans is always available. Non-performing loans is not available for all banks
therefore this paper use the following ratio:
Loan loss reserve to gross loans (LRGL) = Loan loss reserve/gross loans.
This ratio indicates the proportion of the total portfolio that has been set aside but
not charged off. It is a reserve for losses expressed as a percentage of total loans.
5. Results
This section presents and discusses the results.
5.1 Proftability Performance
In banking the risk-reward trade off is constantly present. Risk taking
generates higher expected earnings through various mechanisms. For example
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
41
granting high margin loans to risky customers may increase earnings in the short
term but it also increases the credit risk profle and the probability of future
losses (KPMG, 1998). Figure 1 shows the proftability performance of the South
African commercial banking sector for the period 2005-2009. Proftability is
measured in terms of ROA, ROE, and Cost-to-Income (C/I).
Figure 1: Proftability Trend
Figure 1 shows an increasing proftability trend from 2005 to 2006 with
a slight decrease in 2007. Return on assets increased by 5.14 % from 1.36%
in 2005 to 1.43% in 2006 before slightly falling to 1.40% in 2007. The SARB
believes that the banks’ proftability remained favourable during 2005-2006 due
to strong asset growth as the total assets grew from R1, 474 billion in 2005 to R1,
847 billion in 2006 registering an increase of 25.3%. Loans and advances were
the main contributors to the increase in assets mainly due to increase in mortgage
loans. The higher ratios indicate a better prospective as the high net interest
margin was feeding through greater net income thus boosting ROA and ROE.
However, at the onset of the economic downturn bank performance
deteriorated slightly. ROA decreased from 1.40% for the year 2007 to 1.17%
in 2008 before fnally settling on a low 0.80% for the year 2009, a consequence
of the global fnancial crisis and a slowing down in the domestic economy. The
downward trend is also refected in the ROE, which decreased from 24.01%
in 2007 to 20.33% in 2008 before drastically decreasing to 13.65% in 2009,
refecting a decline of 35.56%. The downward trend is attributable to a decrease
70
60
50
40
30
20
10
0
2005 2006 2007 2008 2009
ROA
ROE
C/I
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
42
in loans and advances to customers as well as increased credit impairments
owing to defaults which negatively impacted proftability.
The trend refected by ROA and ROE is also refected in the cost to income
ratio, which improved by 9.2% from 60.93 in 2005 to 55.32 in 2006 indicating
better effciency and proftability performance. The ratio continued to show signs
of improvement, it strengthened by 6.79% from 51.53 in 2007 to 48.03 in 2008.
The steady improvements in cost to income are mainly ascribed to increasing net
income reported by the banks which rose by 63% from R 17813 million in 2005
to R 29038 million in 2006 consequent of the lower loan loss provisions and
relatively lower operating expenses experienced by the large banks during that
period. The decline in the C/I ratio is as a result of cost effciency levels which
Ncube (2009) in his study of effciency levels in South African banks found to
have signifcantly improved. However, from 2008 the C/I ratio continued to fall
at a decreasing rate and eventually increased to 49.65 in 2009.
5.2 Liquidity Performance
Liquidity performance measures the ability to meet fnancial obligations as
they become due and is crucial to the sustained viability of banking institutions.
What began as credit concerns for the US sub-prime market developed into
concerns in global credit markets with unknown fnancial exposures and
potential losses (ABSA, 2009). The resultant uncertainty made fnancial market
participants exceedingly risk averse, such that they were unwilling to invest in
any markets or fnancial instruments other than ‘safe havens’. This severely
reduced the levels of liquidity in the global fnancial markets (SARB, 2009).
South Africa was not immune to such developments and this is refected in the
liquidity ratios. South African banks rely on customer’s deposits and their current
balances with the South African Reserve bank for their liquidity. These banks are
required to hold an average daily amount of liquid assets that shall not be less
than 5% of adjusted liabilities
3
. Figure 2 shows the liquidity trend in terms of
average net loan to total assets ratio (NLTA), net loan to deposit and short term
borrowing ratio (NLDST), and liquid assets to deposits and short term borrowing
ratio (LADST).
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
43
Figure 2: Liquidity Trend
** Imperial Bank data missing in NLDST and LADST for 2009 at the time of data collection
In as much as the ratio of net loans to total assets does not directly measure
liquidity, it gives an indication of how much of the bank assets are tied into
illiquid loans. From the trend displayed by Figure 2, NLTA increased by 2.84%
from 73.00 in 2005 to 75.08 in 2006 and increased again to 76.48 in 2007 when
favourable economic conditions and preparations for the World Cup 2010
increased the demand for loans from businesses and allowed banks to grow their
loan portfolios. Loans to customers increased by 30% from R1, 005 billion in
2005 to R1, 285 billion in 2006 while total assets increased from R1, 474 billion
in 2005 to R1, 847 billion in 2006 and increased by 21.67% to R2, 247 billion at
the end of 2008.
In 2008 NLTA dropped to 72.94 before fnally increasing again to 73.99 in
2009. The change in the trend signifes the slowing down in loans to customers
and a continued increase in impaired loans leading to a decrease in net loans and
in total assets which consequently decreased by 4.78% to R2, 673 billion at the
end of 2009. Generally, a higher NLTA may indicate possible liquidity problems
for banks in a tight credit market in the face of a large deposit withdrawal or in
case of unexpected withdrawals. However, the increase in NLTA for the fve
banks did not pose any liquidity problems as South African banks could still
access the cash reserves that they held in excess of the minimum requirement
with the reserve bank.
The LADST ratio has been gradually falling for the period under review
NLTA
NLDST
LADST
120
100
80
60
40
20
0
2005 2006 2007 2008 2009
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
44
indicating reduced liquidity for the banks. The ratio decrease by 14.56% from
18.06 in 2005 to 15.43 in 2006 indicating a fall in the amount of customer and
short term funds that could be met if they were suddenly withdrawn. The ratio
however slightly increased by 2.03% in 2008 as the amount of liquid assets
held by the banks increased by 20% during 2009 as banks increased their
investments in instruments qualifying as liquid assets (SARB, 2009). The ratio
then decreased from 15.06 to 13.40 in 2009 indicating a further deterioration in
liquidity. Therefore in as much as banks have been increasing their percentage of
liquid assets that mainly consist of current accounts/reserves with the SARB and
other banks, customer deposits and short term funding have also been increasing
such that the overall trend continued to show reduced levels of liquidity.
NLDST followed a similar trend increasing by 2.16% from 88.02 in 2005
to 89.93 in 2006 and subsequently increasing by 8.55% to 97.62 in 2007. The
increasing trend indicate deteriorating liquidity in the banking sector as more
and more assets, customer deposits and short term funding are tied into loans
which are classifed as illiquid assets. The liquidity of the banks contracted most
between the years 2006-2007 when the banks were aggressively increasing
their loan portfolios during the country’s preparation for World Cup 2010. The
variation in the ratio from 2007 onwards is ascribed to changes in both loans to
customers and changes in deposit and short term funding. The banks tightened
lending standards later in 2008 in response to the global fnancial crisis such that
credit expansion slipped by 2.6% from R1, 842 billion in 2008 to R1, 794 billion
in 2009 eventually reaching very low levels and zero growth by the end of the
year. The lower rate may also be attributed to a tighter monetary policy stance
and the implementation of the stringent risk based lending criteria by the banks
as a result of the introduction of the National Credit Act of 2007. While bank
loans and advances contracted during 2009, the contraction in credit extension
had both demand-side and supply-side elements (SARB, 2009). On the demand
side, it would appear that households continued to be reluctant to incur more debt
leading to a fall in demand for loans, while on the supply side lending standards
have remained tight and led to the slowing down in the growth of loans and
advances to customers.
Financial stability issues lie at the proftability-liquidity nexus therefore
a decline in liquidity is associated with an increase in proftability, since low
liquidity means larger percentage of assets and total deposits are tied with
loans. Under normal circumstances rapid loan growth tend to result in higher
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
45
returns and higher risks. However, Saunders and Cornett (2006) argue that rapid
growth in assets (loans) than in deposits is indicative of banks using borrowed
funds excessively. This seems to highlight the position of the South African
commercial banks. There has been a growing trend in loans and advances to
customers; however, this increase has not been met by an equal increase in
customer deposits. This may then mean South African commercial banks have
been turning to purchased liquidity in the form of money market instruments to
fund the increase in loans. However, this remains uncertain as the data contained
in Bankscope is not adequate to be able to back up the information with detailed
ratios. Furthermore, a closer inspection of the balance sheets shows that retail
deposits represent only about 25% of total deposits in the South African
commercial banking system, while deposits with less than one year maturity
represent close to 80% of total deposits (IMF, 2008).
It is therefore worth noting that as a result of the above mentioned issues
concerning the composition of the deposits, the South African banking system
faces long-standing structural risks rooted in the sector’s reliance on short-term
wholesale deposits. The IMF has recommended that the SARB explore ways
to reduce the risks associated with the banks’ reliance on short term wholesale
deposits. One of the recommendations is implementing a deposit insurance
system to counter such risks and provide the added beneft of inducing household
saving to migrate from unguaranteed liquid fnancial instruments to competing
bank deposits, thus strengthening the retail base of banks.
5.3 Asset Credit Quality (Credit Performance)
Credit performance is concerned with the examination of the risk associated
with a bank’s asset portfolio. Figure 3 shows a fairly stable trend in the loan
reserve to gross loans ratio between 2005 and 2007. However, 2008-2009 shows
a signifcant deterioration in the credit quality. The ratio registered a decline of
12% from 1.83 to 1.61 during 2005-2006 before slightly increasing from 1.61
to 1.68 in 2007. The slight improvement was due to continued growth in loans
to customers as well as growth in non performing loans which continued on a
downward trend.
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
46
Figure 3: Credit Quality Trend
Data for imperial bank for LRGL for 2009 was missing at the time of data collection
The loan portfolio deteriorated in 2008-2009 as the ratio increased by
8.33% from 1.92 in 2008 to 2.08 in 2009. Credit risk ratios increased during
2008 indicating the deterioration of the quality of the loan portfolio as compared
to 2005. For the period 2008 to 2009 nonperforming loans and advances
increased from 2.8% in July 2008, to 5.5 % in August 2009, as a result of the
crisis and the period was also marked by an increase in credit losses in line with
the tougher market conditions. This is because the banks were more exposed to
increased credit risk as risky loans given during the 2005-2006 period began to
go bad and the banks reported higher charge off or additional provision for loan
losses. Hence South Africa’s banks are facing increased credit risk, especially in
their home loan portfolios, in the face of record household indebtedness and a
mounting debt service burden.
Overall, although risk appetites were adjusted in line with challenging
economic conditions and caution was exercised with regard to lending, the
operating environment continued to be under pressure during 2009 as evidenced
by increased credit impairments and resultant lower proftability levels. This
resulted in the quality of the loan portfolio sharply deteriorating during 2008-
2009.
Hypothesis Testing
To examine whether the difference in performance of the banks in 2005-2006
2005 2006 2007 2009 2008
2.50
2.00
1.50
1.00
0.50
0.00
Loan Reserve-Gross loans
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
47
is statistically different from that of 2008-2009 a student’s t- test is employed to
test the hypothesis that the means of the two periods are the same on the seven
variables. The table below provides a summary of a student’s t- test results for
the two periods under review.
Table 2: Student’s T- Test Results
Mean
2005-2006
Mean
2008-2009
P value Decision
Proftability ROA 1.395 0.986 0.006 Reject
ROE 24.149 16.99 0.009 Reject
C/ I 58.125 48.842 0.0009 Reject
Liquidity NLTA 74.04 73.465 0.666 Accept
NLDST 91.686 82.319 0.391 Accept
LADST 16.746 14.226 0.207 Accept
Credit quality LRGR 1.719 1.997 0.542 Accept
With regard to proftability, ROA and ROE shows banks performed better
in the period 2005-2006 compared to 2008-2009. As shown in Table 2, the mean
for ROA was 1.395 for 2005-2006 compared to the 0.986 for 2008-2009. ROE
shows a similar trend with the mean for 2005-2006 being 24.149 compared
to 16.99 for 2008-2009. This indicates that the banks signifcantly progressed
in proftability during 2005-2006. The P- values for ROA and ROE are 0.006
and 0.009 respectively, therefore the differences between the performances for
the two periods are statistically signifcant as the P-values are below 0.05 and
therefore the null hypothesis has to be rejected leading to the conclusion that
proftability deteriorated during 2008-2009. The C/I ratio shows a similar trend,
the difference in the C/I means is statistically signifcant at 95% confdence
level as the P-value is 0.0009 and therefore less than 0.05 the null hypothesis is
therefore rejected. However, in terms of C/I ratio an improved bank performance
is highlighted for the period 2008-2009 as opposed to ROA and ROE which
showed better performance for 2005-2006.
Liquidity levels have been falling as a result of the fnancial crisis. However,
the null hypothesis of equality of means for the two different time periods
cannot be rejected for NLTA and NLDST as the P-values are 0.666 and 0.391
respectively. This implies that statistically, there is no signifcant difference
between the liquidity performance of the banks in the two periods in terms of
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
48
NLTA and NLDST. The means for the LADST shows a different trend with the
mean for 2005-2006 being 16.746 while the one for 2008-2009 is 14.226. This
indicates that banks have been more liquid in 2005-2006 compared to 2008-2009
in terms of LADST. However, with the P-value being 0.207, the differences are
not statistically signifcant. The improved liquidity is a consequence of more
stringent lending procedure and tightening of credit extension procedures in
response to the fnancial crisis.
With respect to credit quality the mean for loan reserve to gross loan
ratio is 1.719 for 2005-2006 and 1.997 for 2008-2009 indicating that the loan
portfolio deteriorated in 2008-2009. However, the difference is not statistically
signifcant as the P-value is 0.542. Therefore, the null hypothesis cannot be
rejected.
From the results of the student t-test, it can be argued that despite the
fnancial turmoil that engulfed the global economy and affected fnancial
institutions around the world, statistically signifcant differences were only
observed in proftability performance of the South African commercial banks.
On the contrary, no signifcant differences were observed between the overall
performances of the commercial banks in South Africa during the two periods
in terms of liquidity and credit quality. This is supported by the null hypothesis
of the equality of the means being accepted on liquidity and credit quality and
rejected on all three proftability ratios. Although the student’s t-test is showing
a mixture of results on the overall performance, the overall results are consistent
with recent literature (See Baxter, 2008; Mminele, 2009) and shows that South
Africa side-stepped the worst of the crisis and has been resilient to the global
fnancial crisis. This is mainly because South African commercial banks had no
direct exposure to the sub-prime mortgage market, while the banks’ international
franchises had very limited exposure (SARB, 2009).
6. Summary and Conclusion
This paper measured the performance of South Africa’s commercial
banking sector over the period 2005-2009. The results indicate that the overall
bank performance in terms of proftability, liquidity, and credit quality has been
improving since 2005 up to and including 2007. Banks increased the size of
their loan portfolios during the period as the country prepared for World Cup
2010. Although the banks aggressively increased their loan portfolios, sound and
effective credit risk management policies have been in place so that the lending
behaviour could still be contained. This is refected in the downward trend in
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
49
nonperforming loans. However, bank performance deteriorated during 2008-
2009 as the banks’ operating environment deteriorated due to the global fnancial
crisis and a slowing economy. The analysis has also uncovered that the illiquidity
levels in the South African commercial banks has reached extreme levels. This
is exacerbated by the banks’ dependence on wholesale markets and the fact that
deposits with less than one year maturity represent close to 80% of total deposits.
Despite these alarming features, South African banks have managed to continue
with their normal day to day business during the global fnancial crisis. South
African banks’ low leverage, high proftability, and limited exposure to foreign
assets and funding allowed them to remain liquid and well-capitalized; obviating
any need for extraordinary liquidity or state support.
We also found signifcant differences in proftability performance
for the period 2005-2006 and the period 2008-2009. The results indicate that
proftability deteriorated during the later period. There might be several reasons
for the signifcant deterioration in proftability. One of the reasons could be
increasing bank operating costs and reduced incomes amid the global fnancial
crisis. Furthermore in these recessionary times, when corporate and private
clients fnd it hard to service their debts, the level of the provision for loan losses
and bad debts increased. In contrast, no statistically signifcant differences were
observed between bank performance during the two periods in terms of liquidity
and credit quality. The comparable performance results in terms of liquidity
and credit quality between these two periods may be as a result of the fact that
South Africa entered the downturn with a sound macro/fscal position, enabling
aggressive counter-cyclical fscal and monetary responses. Notwithstanding the
turmoil experienced in international fnancial markets and the domestic cyclical
economic developments during 2008-2009, the South African banking system
remained stable; banks were adequately capitalised and proftable. South African
banks remained in a sound position weathering the global fnancial storm, as they
benefted from limited exposure to foreign currency debt and the fact that assets
at the epicentre of the crisis were minimal in South Africa (SARB, 2008).
Notes:
1http://web.worldbank.org/WBSITE/EXTERNAL/COUNTRIES/
AFRICAEXT/SOUTHAFRICAEXTN
2.
Investec Bank has been omitted from the sample as it is an investment bank
rather than a commercial bank. Capitec Bank, Teba Bank and African bank
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
50
have been excluded on the basis that they are micro fnancing institutions
with different profles from the fve banks in the sample.
3. Adjusted liabilities’ refers to total liabilities reduced by (1) funding received
from head offce or from other branches within the same group; and (2)
amounts owing by banks, branches and mutual banks in South Africa.
References
Ahmed, M., B. (2009). Measuring the Performance of Islamic Banks by
Adapting Conventional Ratios German University in Cairo Faculty of
Management Technology Working Paper No. 16 pp 1-26.
Baxter, R. (2008). The global economic crisis and its impact on South Africa
and the country’s mining industry South African Reserve bank conference
papers [Online] Available at:http://www.reservebank.co.za/internet/publication.nsf/WCEV/498F27F900B18
F27422576010035EE92/?opendocument [Accessed 14 June 2010]
Berger, A.N. and Humphrey, D.B. (1997). Effciency of fnancial institutions:
international survey and directions for future research. European Journal of
Operational Research, Vol 98, pp.175-212.
Brand, A (2009). Lagos is not Lehman‘– The Differentiated Impact of the
Global Financial Crisis on Emerging Markets and Developing Countries.
University of Economics, Prague, Publishing house 1st Edition Volume 2
pp. 73-103.
BUREU VAN DIJK (2009). Bankscope Database.
Calem, P. and Carlino, G. (1991) The Concentration/Conduct Relationship in
Bank
Deposit Markets, Review of Economics and Statistics 73 pp. 268-76.
Casu, B., Molyneux, P. And Girardone, C. (2006). Introduction to Banking,
Prentice Hall/ Financial Times, London.
Cronje, T. (2007). Assessing the relative effciency management of South
African banks. Management Dynamics 16 pp 11-23.
Dietrich, J. (1996). Financial Services and Financial Institutions: Value
Creation in Theory and Practice Prentice Hall.
Drake, L. (2001). Effciency and productivity change in UK banking. Applied
Financial Economics 11, pp 557-571.
Gilbert, E., Calitz, E., Plessis, S., (2009). Prudential regulation, its international
background and the performance of the banks a critical review of the South
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
51
African environment since 1970. Economic History of Developing Regions
24 (2), pp 43 — 81.
Greenberg, B. J., & Simbanegavi, W. (2009). Testing for completion in South
Africa Banking Sector Unpublished Manuscript.
Halkos, G. and Salamouris, D. (2004). Effciency measurement of the Greek
commercial banks with the use of fnancial ratios: a data envelope analysis
approach. Management Accounting Research 15 (2), pp 201- 224.
Hempel, G., Simonson, D., and Coleman, A. (1994). Bank Management: Text
and Cases. 4th Edition, John Wiley & Sons, Inc.
IMF (2008). South Africa: Financial System Stability Assessment. IMF
Country Report No. 08/349 [Online] Available onhttp://www.imf.org/
external/pubs/ft/scr/2008/cr08349.pdf [Accessed 19 July 2010]
Kirkpatrick, C., Murinde, V. and Tefula, M. (2007).The measurement and
determinants of x-ineffciency in commercial banks in Sub-Saharan Africa.
European Journal of Finance 14 (7) 2, pp 625-639.
Kiyota, H. (2009).Effciency of Commercial Banks in Sub-Saharan Africa: A
Comparative Analysis of Domestic and Foreign Banks. A paper prepared for
the CSAE conference 2009 on “Economic Development in Africa” held at
the University of Oxford.
KPMG (1998). Bank margins and related proftability in South Africa A study.
[Online], Available on:http://www.scribd.com/doc/14752909/Banking-Ass-
KPMG-Study-Bank-Margins [Accessed 01 August 2010].
Lacewell, S., K. (2003). Do Effcient Institutions Score Well Using Ratio
Analysis? An Examination Of Commercial Banks in The 1990s. Journal of
Commercial Banking and Finance 2, pp 17-33.
Mboweni, T., .T (2004). The South African banking sector - an overview of the
past 10 years. Address by Governor of the South African Reserve Bank, at
the Year-end media cocktail function, Johannesburg, 14 December 2004.
Mminele A.D (2009). Recent economic developments in South Africa:
Citigroup Global Issues Seminar Istanbul 4 October 2009 [Online]
Available onhttp://www.bis.org/review/r091008d.pdf [Accessed 10 August
2010].
Murinde, V. (2009). Global Financial Crisis: Implications for Africa’s Financial
System, paper prepared for the European Development Report (ERD) 2009
Conference on “Financial markets adverse shocks and policy responses in
fragile countries”, 21-23 May, Accra, Ghana.
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
52
Napier, M. (2006). Provision for Financial Services in South Africa in
Liberalisation and universal access to basic services: telecommunications,
Water and Sanitation, Financial Services, and Electricity OECD policy
studies and The World Bank Published by: OECD Publishing.
Ncube, M. (2009). Effciency of the Banking Sector in South Africa, African
Economic Conference 2009 Fostering Development in an Era of Financial
and Economic Crises, Addis Ababa.
Oberholzer, M. and Van der Westhuizen, G. (2004). An Empirical Study
on Measuring Effciency and Proftability of Bank Regions Meditari
Accountancy Research 12 (1), pp 165–178.
O’Donnell, C.J. and van der Westhuizen, G. (2002). Regional comparisons
of banking performance in South Africa. The South African Journal of
Economics 70 (3), pp 485-518.
Okeahalam, C., C. (2006). Production costs and effciency in the South African
banking sector: a stochastic analysis. International Review of Applied
Economics 20, pp 103–23.
Ramanathan, R. (2007). Performance of Banks in Countries of the Gulf
Cooperation Council. International Journal of Productivity and
Performance Management 56 (2), pp 137-154.
Ramcharan, R. (2009). Sound Policies Shield South Africa from Worst of
Recession. IMF African Department IMF survey magazines: countries and
regions survey.
Rossouw, J. (2009). South African Reserve Bank History, functions and
institutional structure SARB [Online] Available onhttp://www.reservebank.
co.za [Accessed 08 June 2010].
Samad, A. (2004). Bahrain Commercial Bank’s Performance during 1994-2001.
Credit and Financial Management Review 10(1) pp 33-40.
Saunders, A and Cornett, M., M. (2006). Financial institutions management: a
risk management approach 5
th
Edition McGraw-Hill.
Seiford, L., M and Zhu, J (1999). Proftability and Marketability of the Top 55
U.S. Commercial Banks Management Science 45 (9), pp 1270-1288.
Sinkey, J. (2002). Commercial Bank Financial Management: In the Financial-
Services Industry 6
th
Edition Prentice Hall, Upper Saddle River, NJ.
South African Reserve Bank (2008). Financial Stability Review, Pretoria: South
African Reserve Bank. [Online] available onhttp://www.reservebank.co.za/
[accessed 15 June 2010].
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
53
South African Reserve Bank Supervision Department Annual Report 2005,
2006; 2007; 2008; 2009 Pretoria [Online] available onhttp://www.
reservebank.co.za/ [accessed 15 June 2010].
Tarawneh, M. (2006). A Comparison of Financial Performance in the Banking
Sector: Some Evidence from Omani Commercial Banks”. International
Research Journal of Finance and Economics 3, pp 103-112.
Thygerson, K., J. (1995). Management of fnancial Institutions, HarperCollins
College Publishers.
Van der Walt S.J. (1998). “The Infuence of Socio-Political Developments on
Banks in South Africa, 1992 to 1997”, Occasional Paper No.13, South
African Reserve
Bank, Pretoria, RSA
Webb, R.M. (2003). Levels of effciency in UK retail banks: a DEA window
analysis. International Journal of the Economics of Business Vol 10 (3), pp
305-322.
World Bank South Africa country Brief [Online] Available onhttp://web.
worldbank.org/WBSITE/EXTERNAL/COUNTRIES/AFRICAEXT/
SOUTHAFRICAEXTN [Accessed 24 July 2010].
Yeh, Q., J. (1996). The Application of Data Envelopment Analysis in
Conjunction with Financial Ratios for Bank Performance Evaluation.
Journal of the Operational Research Society 47, pp 980-988.
doc_422527917.pdf
This paper investigates the performance of South Africa's commercial banking sector for the period 2005- 2009. Financial ratios are employed to measure the profitability, liquidity and credit quality performance of five large South African based commercial banks.
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
30
African Review of Economics and Finance, Vol. 2, No. 1, Dec 2010
©The Author(s)
Journal compilation ©2010 African Centre for Economics and Finance. Published by
Print Services, Rhodes University, P.O.Box 94, Grahamstown, South Africa
A fnancial Ratio Analysis of Commercial Bank
Performance in South Africa
Mabwe Kumbirai
2#
and Robert Webb*
Abstract
This paper investigates the performance of South Africa’s commercial banking
sector for the period 2005- 2009. Financial ratios are employed to measure
the proftability, liquidity and credit quality performance of fve large South
African based commercial banks. The study found that overall bank performance
increased considerably in the frst two years of the analysis. A signifcant change
in trend is noticed at the onset of the global fnancial crisis in 2007, reaching its
peak during 2008-2009. This resulted in falling proftability, low liquidity and
deteriorating credit quality in the South African Banking sector.
1. Introduction
Commercial banks in South Africa have undergone immense regulatory and
technological changes since the attainment of constitutional democracy in 1994.
South African banks are faced with increasing competition and rising costs as a
result of regulatory requirements, fnancial and technological innovation, entry
of large foreign banks in the retail banking environment and challenges of the
recent fnancial crisis. These changes had a dramatic effect on the performance
of the commercial banks. Most studies on bank performance in South Africa
have focused on branch performance [see Oberholzer and Van der Westhuizen
(2004); O’Donnell and Van der Westhuizen, (2002); Okeahalam (2006)]. More
recently, Cronje (2007) and Ncube (2009) studied the effciency of South
African banks using Data Envelopment Analysis (hereafter DEA), studying
the periods 1997-2007 and 2000-2005 respectively. This study evaluates bank
2 Caledonian Business School, Glasgow Caledonian University, Cowcaddens Road G4 0BA,
UK
#
Corresponding author: [email protected]
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
31
performance for the period 2005-2009 using fnancial ratio analysis (hereafter
FRA). The present study is different from earlier studies in two ways: sample
coverage and methodology. Covering South Africa’s big banks in the period both
prior to, and after the 2007 subprime meltdown highlights important changes
that have occurred in the banking industry and tease out appropriate policy for
improving bank performance. Compared to extant literature we favour FRA
because it is effective in distinguishing high performing banks from others, tends
to compensates for disparities and controls for any size effect on the fnancial
variables being studied (Samad, 2004). Additionally, fnancial ratios enable us
to identify unique bank strengths and weaknesses, which in itself inform bank
proftability, liquidity and credit quality. The rest of the paper is organised as
follows: the next section offers background information on the fnancial system
in South Africa. Section 3 outlines the past studies in bank performance. The
methodology and data used are described in section 4. Section 5 presents and
analyses the results and section 6 concludes.
2. Banking in South Africa
While ranked in the top 20 of world economies by size, the South African
economy remain relatively small accounting for less than 1 per cent of global
GDP (Baxter, 2008). Despite being small by global standards, South Africa is
the economic powerhouse of Africa, leading the continent in industrial output
and mineral production and generating a large proportion of Africa’s electricity
(Brand, 2009). South Africa’s economic performance during the frst decade
of freedom was impressive, with favourable external environment and strong
domestic demand helping raise GDP growth to 5 % on average in 2004–2007 and
lowering the unemployment rate by 5% (Ramcharan, 2009).
The consistent performance ensured by a prolonged economic expansion,
supported by prudent economic policies and improving macroeconomic
fundamentals resulted in low infation, high commodity prices and increased
investor confdence. In such favourable economic conditions, the banking sector
played an essential role in the economic growth of the country. However, the
global fnancial crisis of late 2007 sharply changed the outlook for an already
slowing economy and South Africa was not immune to the impact of the global
fnancial crisis-induced economic slowdown (Baxter, 2008). Slower economic
growth in key export markets, lower commodity prices and a slowdown in
capital fows to developing countries had a negative impact on the South African
economy. Negative growth was experienced in the fourth quarter of 2008 and
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
32
the South African economy offcially entered into a recession in the frst quarter
of 2009 (SARB, 2009). The table below shows the South African economic
indicators for the period 2005-2009. The real GDP growth has been increasing
since 2000. The growth rate was 5.3% for 2006 before falling to 5.1% in 2007
when the fnancial crisis started. However as can be seen, the growth rate then
drastically fell to 3.1% for 2008 before settling on a negative 1.8 % in 2009
refecting the above mentioned effects of the fnancial crisis on the South African
economy GDP growth. Infation, which has remained outside the 3–6 % target
band since 2007, peaked to 11.5% in 2008 as shown in table 1.
Table 1: South Africa’s Key Economic Indicators
2005 2006 2007 2008 2009
Real GDP 5.0 5.3 5.1 3.1 -1.8
CPI (annual average) 3.4 4.7 7.1 11.5 7.4
CPIX (end of period) 1 4.0 5.0 8.6 10.3 Na
Unemployment rate (percent) 26.2 25.5 22.7 21.9 24.9
Broad money (end of period) 20.5 22.5 23.6 14.8 7.9
National government budget balance
(percent of GDP)
-0.6 0.4 0.8 -0.7 -4.4
National government debt (percent of
GDP)
35.2 33.0 28.5 27.3 30.2
External current account balance
(percent of GDP)
-4.0 -6.3 -7.3 -7.4 -4.4
External debt (percent of GDP) 19.0 22.1 26.6 25.9 27.3
Gross reserves 2.9 3.1 3.7 4.6 4.5
US Dollar exchange rate (rand per u.s.
dollar
6.33 6.97 6.81 9.53 Na
Sources: South African Reserve Bank; IMF, International Financial Statistics; and
IMF Staff projections. Since January 2009, a reweighed and rebased CPI replaced the
previously used CPIX (the consumer price index excluding the interest on mortgage
loans) as the targeted measure of infation.
Despite the apparent worsening of the economic indicators, South Africa
maintained strong macro-fscal stabilization measures and had progressed on
several socio-economic fronts before the crisis
1
hence the economy’s resilience
to the global fnancial crisis. Consequently, for an open economy intricately
woven into the ‘fabric’ of the global economy, South Africa weathered the global
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
33
crisis relatively well compared to other countries in the region (Baxter, 2008).
South Africa’s 1994 transition from apartheid to constitutional democracy
has been one of the most important political achievements of its time. The safe
political and business environment resulting from the attainment of democracy
opened up opportunities for new domestic entrants to the banking system (Van
der Walt, 1998). The biggest increase in new bank registrations was in 1996,
when registrations rose from 35 to 44. However, this was followed by the biggest
shakeout during 1999 when these banks faced liquidity pressures which saw them
exiting the banking system (Mboweni, 2004). Between 1996 and 2004 more than
half of these banks disappeared, including Saambou in 2002 when, like a few
other small banks, it was denied access to lender-of last resort facilities (Gilbert
et al, 2009). Mboweni, (2004) has argued that the downward spiralling of the
small and medium banks was due to consolidation in the banking sector rather
than failure of the small and medium banks. A fact also supported by Gilbert
et al, (2009:68) who states that the change in the banking scene was as a result
of the ‘‘rationalisation of the banking system whereby organisational specifc,
functional specialisation was replaced by functional diversifcation within
multifunction institutions’’.
The socio-political changes led to structural changes in the banking sector as
well as the delivery of fnancial services. The political transformation, relaxation
of exchange controls and liberalisation of the economy resulted in South Africa
increasingly becoming an important fnancial centre (Mboweni, 2004; Murinde,
2009). Such fnancial liberalisation did not only bring greater competition for
traditional domestic banks but also improved the quality and availability of
fnancial services in South Africa. In September 2004 Barclays announced its
intention to acquire a majority shareholding in ABSA, South Africa’s second
largest bank. This event signalled the UK banking giant’s return to banking
in South Africa having been forced out by economic sanctions and political
instability in 1986. These events signifed the reintegration of the South African
domestic economy into the global environment evidenced by the relaxation of
the exchange controls and the provision of fnancial services to the poor, non-
white population that was practically excluded from the formal fnancial services
in the apartheid era (Napier, 2006).
The South African banking industry is also characterised by international
links through correspondent banking relations with off-shore banks and
international institutional investments in domestic banks. Therefore, regulation
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
34
and supervision of banking is still an issue, especially with respect to fnancial
innovations such as derivative instruments and in light of the 2007-2009 global
fnancial crisis (Murinde, 2009). The 1990 Banks Act was amended in 2008,
to align the South African banking legislative framework with the principles of
Basel II. Banking regulation is aimed at a balance between enhanced stability of
the banking sector and the costs of reduced competition, struck by successive
rounds of bank regulation (Gilbert et al, 2009). Generally, South Africa’s
approach to fnancial system stability places considerable reliance on private
and market forces to achieve fnancial system stability with Rossouw (2009)
arguing that any government intervention would be at the minimum level and
only needed to contain systemic risk.
At the heart of the South African banking system is the South African
Reserve Bank (SARB), established in 1921. As at the end of 2009 the banking
sector consisted of 12 locally controlled banks, 6 foreign controlled banks and
2 mutual banks (SARB, 2009). Further, the banking sector has 42 international
banks with authorised representative offces in South Africa. The banking
sector is highly concentrated with the largest four banks holding 86.4% of the
total industry assets (SARB, 2009; Greenberg and Simbanegavi, 2009). Most
economists would analyse the level of concentration in a banking industry using
indicators such as Herfndahl–Hirschman Index, N-frm concentration ratio and
Learner’s index of monopoly (Calem and Carlino, 1991). An H-index above 0,
18 represents a highly concentrated industry that goes some way to indicating
the presence of an oligopoly. An ‘oligopoly’ can be defned as an imperfectly
competitive market structure in which a few institutions dominate the industry.
Such a description may ft the South African banking industry whose index has
been above 0.18 since 2005 (SARB, 2009).
Although the South African banking sector has been relatively insulated
from the direct impacts of the global fnancial-sector crisis through appropriate
monitoring and supervision of the domestic banking sector, the negative
contagion effects of the crisis had a negative effect on bank balance sheets
(SARB, 2008: 2009). The aggregated balance sheet of the banking sector in
South Africa equalled R1 677 billion in 2005. The sector’s balance-sheet size
then grew to R3 177 billion 2008 (135, 4 % of GDP), followed by a decline
in asset growth during 2009, ending the year at R2 967 billion (118, 5 % of
GDP). Banking sector assets comprise mainly loans and advances, followed
by derivative fnancial instruments. Home loans and term loans represent
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
35
approximately 52% of the total assets while commercial mortgages represent
9.7%. On the liabilities side, deposits constitute a signifcant percentage of
banking-sector liabilities amounting to 79, 6 % in 2008 and 85, 4 % in 2009.
Deposits by corporate customers which constitute the largest portion of banking-
sector deposits amounted to 42, 5 % in 2009, followed by retail customers and
bank deposits, which accounted for 22, 3 % and 13, 7 %.
3. Previous research on bank performance
The measurement of bank performance particularly commercial banks
is well researched and has received increased attention over the past years
(Seiford and Zhu, 1999). There have been a large number of empirical studies
on commercial bank performance around the world (see Yeh, 1996; Webb, 2003;
Lacewell, 2003; Halkos and Salamouris, 2004; Tarawneh, 2006). However,
little has been done on bank performance in South Africa. However, with the
deteriorating health of the banking institutions and the recent surge of bank
failures as a result of the current global fnancial crisis, it is justifed that bank
performance receives increased investigation from both scholars and industry
specialists.
There are two broad approaches used to measure bank performance, the
accounting approach, which makes use of fnancial ratios and econometric
techniques. Traditionally accounting methods primarily based on the use of
fnancial ratios have been employed for assessing bank performance (Ncube,
2009). However, the limitations of this method coupled with advances in
management sciences have led to the development of alternate methods such
as non-parametric DEA and parametric Stochastic Frontier Approach (hereafter,
SFA) (Berger and Humphrey, 1997).
Berger & Humphrey (1997) assert that the whole idea of measuring bank
performance is to separate banks that are performing well from those which
are doing poorly. They further indicated that, “evaluating the performance of
fnancial institution can inform government policy by assessing the effects of
deregulation, mergers and market structure on effciency” (p175). Bank regulators
screen banks by evaluating banks’ liquidity, solvency and overall performance
to enable them to intervene when there is need and to gauge the potential for
problems (Casu et al, 2006). On a micro-level, bank performance measurement
can also help improve managerial performance by identifying best and worst
practices associated with high and low measured effciency.
This brief review will focus upon studies on South Africa and other
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
36
emerging economies. When looking to improve their performance, banks
compare the performance of their peers and evaluate the trend of their fnancial
performance over time. Tarawneh (2006) in his study measured the performance
of Oman commercial banks using fnancial ratios and ranked the banks based
on their performance. The study utilised FRA to investigate the impact of asset
management, operational effciency and bank size on the performance of Oman
commercial banks. The fndings indicated that bank performance was strongly
and positively infuenced by operational effciency, asset management and bank
size.
In the Gulf, Samad (2004) investigated the performance of seven locally-
incorporated commercial banks during the period 1994-2001. Financial ratios
were used to evaluate the credit quality, proftability, and liquidity performances.
The performance of the seven commercial banks was compared with the
banking industry in Bahrain which was considered a benchmark. The article
applied a Student’s t-test to measure the statistical signifcance for the measures
of performance. The results revealed that commercial banks in Bahrain were
relatively less proftable, less liquid and were exposed to higher credit risk than
the banking industry, in which wholesale banks are the main component.
Kiyota (2009) in a two- stage procedure investigated the proft effciency
and cost effciency of commercial banks operating in 29 Sub-Saharan African
countries during 2000-2007. The article employs the SFA for the estimation of
proft and cost effciency, fnancial ratios and the Tobit regression to provide
cross-country evidence on the performance and effciency of African commercial
banks. The fndings based on a range of performance ratios as well as stochastic
cost and proft frontier estimation, suggest that foreign banks tend to outperform
domestic banks in terms of proft effciency as well as cost effciency. The results
are also in line with the research by Kirkpatrick et al (2007) who used a sample of
89 banks from Sub-Sahara African countries for the period 1992-1999 and found
that banks are on average 67% proft effcient and 80% cost effcient, as indicated
by the results from both the distribution free approach and SFA methods.
O’Donnell and Van der Westhuizen (2002) measured the effciency of a
South African bank at branch level. Their main focus was investigating branches
which were performing well and those that were doing badly, where effciency
could be improved. They found that many branches were operating on a scale
that is too small and could increase their operational scales thereby improving
the overall effciency of the bank. In a similar approach, Okeahalam (2006)
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
37
used the Bayesian SFA to assess the production effciency of 61 bank branches
in nine provinces of South Africa. The fndings of the study points to the fact
that although every branch is operating at increasing returns to scale, the bank
branches can reduce their cost by 17% if they improve the level of effciency.
Overall the article concludes that the bank branches are less effcient than they
should be and could obtain cost reductions by increasing output.
Further, Oberholzer and Van der Westhuizen (2004) investigated the
effciency and proftability of ten banking regional offces of one of South
Africa’s larger banks. This study demonstrates how conventional proftability
and effciency analyses can be used in conjunction with DEA. Although their
study concentrated on banking regions, their fndings confrm those of Yeh
(1996) that DEA results as an effciency measure have a relationship with both
proftability and effciency ratios. The conclusions were that there are signifcant
relationships between conventional proftability and effciency measures and
allocative, cost and scale effciency and no signifcant relationship with technical
effciency.
Most of the above mentioned studies concentrated on branches of a single
bank, one of the studies that investigated the entire South African banking
sector is Cronje (2007) who employed the DEA method and a sample of 13
South African banks to provide a measure of the effciency of the South African
banks. His fndings show that out of the 13 banks, the three largest banks are
effcient and serve as a standard for the banks classifed as ineffcient. The fourth
largest bank showed a slight ineffciency. Overall, seven banks were classifed as
ineffcient and the article recommends target areas for the banks to improve their
effciencies with guidelines that bankers in ineffcient banks could use to increase
their sustainable proftability. The results of this study are in sharp contrast to
studies in the UK where Drake (2001) and Webb (2003) found the larger banks
less effcient. This difference could be attributed to the differences in operating
environment as South Africa is an emerging economy with a different political
and economic history where as UK is a developed country.
Another study that provides a brief but interesting account of bank
performance was conducted by Ncube (2009) who uses the stochastic frontier
model to analyse the cost and proft effciency of four large and four small South
African banks. The results of the study show that South African banks have
signifcantly improved their cost effciencies between 2000 and 2005 with the
most cost effcient banks also being most proft effcient. However, effciency
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
38
gains on proftability over the same time period were found not to be signifcant.
4. Methodology and Data
This paper uses a descriptive fnancial ratio analysis to measure, describe and
analyse the performance of commercial banks in South Africa during the period
2005-2009. Additionally, to examine whether the difference in performance of
the banks in 2005-2006 is statistically different from that of 2008-2009 a student’s
t- test is employed to test the hypothesis that the means of the two periods are the
same on the seven variables as detailed in section 4.1.
The following hypothesis has been tested:
H
0:
µ
1
=µ
2
,
Where µ
1
is the mean for 2005-2006 and µ
2
is the mean for 2008-2009.
Inferences about the hypothesis are made by looking at test statistics and critical
values associated with the mean. If P-value ? ?, reject the null hypothesis. If
P-value > ?, do not reject the null hypothesis. The results of the test are to be
handled with caution as there are very few observations and the statistical tool
might not be very effective when the sample is small.
The selection of the FRA method for this study is motivated by the fact
that from the review of past studies on banking in South Africa and to the
researchers’ knowledge, no researcher has used FRA to measure the performance
of commercial banks in South Africa during 2005-2009. Authors Oberholzer and
Van der Westhuizen (2004) used the method in measuring branch performance
however, the authors concentrated on branches of a single large bank. The
main advantage of FRA is its ability and effectiveness in distinguishing high
performance banks from others and the fact that FRA compensates for disparities
and controls for any size effect on the fnancial variables being studied (Samad,
2004). Additionally, fnancial ratios can be used to identify a bank’s specifc
strengths and weaknesses as well as providing detailed information about bank
proftability, liquidity and credit quality policies (Hempel et al, 1994: Dietrich,
1996). FRA permits a historical sketch of bank returns and risks which Hempel
et al, (1994) suggests presents an opportunity to evaluate the past performance
of the bank which is an important step for planning for future performance.
Although accounting data in fnancial statements is subject to manipulation and
fnancial statements are backward looking, they are the only detailed information
available on the bank’s overall activities (Sinkey, 2002). Furthermore, they are
the only source of information for evaluating the management’s potential to
generate satisfactory returns in future.
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
39
The population for this research comprise of all the banks operating in
South Africa between 2005 and 2009. A sample of the top fve commercial
banks was selected based on the value of their total assets at the end of the
2009 fnancial year end. These are the banks that dominate the sector with the
top 4 banks controlling 84.6% of the total industry assets which makes them
systemically important banks. The fact that all banks could not be included in the
study constrains the validity of the study. However, only the fve largest South
African banks (FirstRand Bank, Absa, Nedbank, Standard Bank and Imperial
bank
2
) offer a comprehensive variety of fnancial services right through South
Africa, all the other banks are aimed at niche markets or confned to geographical
operations. The data was obtained from Bankscope and the bank’s fnancial
statements and websites.
4.1 The Variables
A. Proftability Performance
The most common measure of bank performance is proftability. Proftability is
measured using the following criteria:
Return on Assets (ROA) = net proft/total assets shows the ability of
management to acquire deposits at a reasonable cost and invest them in proftable
investments (Ahmed, 2009). This ratio indicates how much net income is
generated per £ of assets. The higher the ROA, the more the proftable the bank.
Return on Equity (ROE) = net proft/ total equity. ROE is the most important
indicator of a bank’s proftability and growth potential. It is the rate of return to
shareholders or the percentage return on each £ of equity invested in the bank.
Cost to Income Ratio (C/I) = total cost /total income measures the income
generated per £ cost. That is how expensive it is for the bank to produce a unit of
output. The lower the C/I ratio, the better the performance of the bank.
B. Liquidity performance
Liquidity indicates the ability of the bank to meet its fnancial obligations in a
timely and effective manner. Samad (2004:36) states that ‘‘liquidity is the life
and blood of a commercial bank’’. Financial liabilities are attracted through
retail and wholesale distribution channels. Retail generated funding is
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
40
considered less interest elastic and more reliable than deposits attracted
from wholesale distribution channels (Thygerson, 1995). The following ratios
are used to measure liquidity.
Liquid assets to deposit-borrowing ratio (LADST) = liquid asset/customer
deposit and short term borrowed funds. This ratio indicates the percentage of
short term obligations that could be met with the bank’s liquid assets in the case
of sudden withdrawals.
Net Loans to total asset ratio (NLTA) = Net loans/total assets NLTA measures
the percentage of assets that is tied up in loans. The higher the ratio, the less
liquid the bank is.
Net loans to deposit and borrowing (NLDST) = Net loans/total deposits and
short term borrowings. This ratio indicates the percentage of the total deposits
locked into non-liquid assets. A high fgure denotes lower liquidity.
C. Asset Credit Quality (Credit Performance)
While it is expected that banks would bear some bad loans and losses in their
lending activities, one of the key objectives of the bank is to minimize such
losses (Casu et al, 2006). Credit performance evaluates the risks associated with
the bank’s asset portfolio i.e. the quality of loans issued by the bank. Several
ratios can be used for measuring credit quality however, not all information on
the loans is always available. Non-performing loans is not available for all banks
therefore this paper use the following ratio:
Loan loss reserve to gross loans (LRGL) = Loan loss reserve/gross loans.
This ratio indicates the proportion of the total portfolio that has been set aside but
not charged off. It is a reserve for losses expressed as a percentage of total loans.
5. Results
This section presents and discusses the results.
5.1 Proftability Performance
In banking the risk-reward trade off is constantly present. Risk taking
generates higher expected earnings through various mechanisms. For example
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
41
granting high margin loans to risky customers may increase earnings in the short
term but it also increases the credit risk profle and the probability of future
losses (KPMG, 1998). Figure 1 shows the proftability performance of the South
African commercial banking sector for the period 2005-2009. Proftability is
measured in terms of ROA, ROE, and Cost-to-Income (C/I).
Figure 1: Proftability Trend
Figure 1 shows an increasing proftability trend from 2005 to 2006 with
a slight decrease in 2007. Return on assets increased by 5.14 % from 1.36%
in 2005 to 1.43% in 2006 before slightly falling to 1.40% in 2007. The SARB
believes that the banks’ proftability remained favourable during 2005-2006 due
to strong asset growth as the total assets grew from R1, 474 billion in 2005 to R1,
847 billion in 2006 registering an increase of 25.3%. Loans and advances were
the main contributors to the increase in assets mainly due to increase in mortgage
loans. The higher ratios indicate a better prospective as the high net interest
margin was feeding through greater net income thus boosting ROA and ROE.
However, at the onset of the economic downturn bank performance
deteriorated slightly. ROA decreased from 1.40% for the year 2007 to 1.17%
in 2008 before fnally settling on a low 0.80% for the year 2009, a consequence
of the global fnancial crisis and a slowing down in the domestic economy. The
downward trend is also refected in the ROE, which decreased from 24.01%
in 2007 to 20.33% in 2008 before drastically decreasing to 13.65% in 2009,
refecting a decline of 35.56%. The downward trend is attributable to a decrease
70
60
50
40
30
20
10
0
2005 2006 2007 2008 2009
ROA
ROE
C/I
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
42
in loans and advances to customers as well as increased credit impairments
owing to defaults which negatively impacted proftability.
The trend refected by ROA and ROE is also refected in the cost to income
ratio, which improved by 9.2% from 60.93 in 2005 to 55.32 in 2006 indicating
better effciency and proftability performance. The ratio continued to show signs
of improvement, it strengthened by 6.79% from 51.53 in 2007 to 48.03 in 2008.
The steady improvements in cost to income are mainly ascribed to increasing net
income reported by the banks which rose by 63% from R 17813 million in 2005
to R 29038 million in 2006 consequent of the lower loan loss provisions and
relatively lower operating expenses experienced by the large banks during that
period. The decline in the C/I ratio is as a result of cost effciency levels which
Ncube (2009) in his study of effciency levels in South African banks found to
have signifcantly improved. However, from 2008 the C/I ratio continued to fall
at a decreasing rate and eventually increased to 49.65 in 2009.
5.2 Liquidity Performance
Liquidity performance measures the ability to meet fnancial obligations as
they become due and is crucial to the sustained viability of banking institutions.
What began as credit concerns for the US sub-prime market developed into
concerns in global credit markets with unknown fnancial exposures and
potential losses (ABSA, 2009). The resultant uncertainty made fnancial market
participants exceedingly risk averse, such that they were unwilling to invest in
any markets or fnancial instruments other than ‘safe havens’. This severely
reduced the levels of liquidity in the global fnancial markets (SARB, 2009).
South Africa was not immune to such developments and this is refected in the
liquidity ratios. South African banks rely on customer’s deposits and their current
balances with the South African Reserve bank for their liquidity. These banks are
required to hold an average daily amount of liquid assets that shall not be less
than 5% of adjusted liabilities
3
. Figure 2 shows the liquidity trend in terms of
average net loan to total assets ratio (NLTA), net loan to deposit and short term
borrowing ratio (NLDST), and liquid assets to deposits and short term borrowing
ratio (LADST).
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
43
Figure 2: Liquidity Trend
** Imperial Bank data missing in NLDST and LADST for 2009 at the time of data collection
In as much as the ratio of net loans to total assets does not directly measure
liquidity, it gives an indication of how much of the bank assets are tied into
illiquid loans. From the trend displayed by Figure 2, NLTA increased by 2.84%
from 73.00 in 2005 to 75.08 in 2006 and increased again to 76.48 in 2007 when
favourable economic conditions and preparations for the World Cup 2010
increased the demand for loans from businesses and allowed banks to grow their
loan portfolios. Loans to customers increased by 30% from R1, 005 billion in
2005 to R1, 285 billion in 2006 while total assets increased from R1, 474 billion
in 2005 to R1, 847 billion in 2006 and increased by 21.67% to R2, 247 billion at
the end of 2008.
In 2008 NLTA dropped to 72.94 before fnally increasing again to 73.99 in
2009. The change in the trend signifes the slowing down in loans to customers
and a continued increase in impaired loans leading to a decrease in net loans and
in total assets which consequently decreased by 4.78% to R2, 673 billion at the
end of 2009. Generally, a higher NLTA may indicate possible liquidity problems
for banks in a tight credit market in the face of a large deposit withdrawal or in
case of unexpected withdrawals. However, the increase in NLTA for the fve
banks did not pose any liquidity problems as South African banks could still
access the cash reserves that they held in excess of the minimum requirement
with the reserve bank.
The LADST ratio has been gradually falling for the period under review
NLTA
NLDST
LADST
120
100
80
60
40
20
0
2005 2006 2007 2008 2009
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
44
indicating reduced liquidity for the banks. The ratio decrease by 14.56% from
18.06 in 2005 to 15.43 in 2006 indicating a fall in the amount of customer and
short term funds that could be met if they were suddenly withdrawn. The ratio
however slightly increased by 2.03% in 2008 as the amount of liquid assets
held by the banks increased by 20% during 2009 as banks increased their
investments in instruments qualifying as liquid assets (SARB, 2009). The ratio
then decreased from 15.06 to 13.40 in 2009 indicating a further deterioration in
liquidity. Therefore in as much as banks have been increasing their percentage of
liquid assets that mainly consist of current accounts/reserves with the SARB and
other banks, customer deposits and short term funding have also been increasing
such that the overall trend continued to show reduced levels of liquidity.
NLDST followed a similar trend increasing by 2.16% from 88.02 in 2005
to 89.93 in 2006 and subsequently increasing by 8.55% to 97.62 in 2007. The
increasing trend indicate deteriorating liquidity in the banking sector as more
and more assets, customer deposits and short term funding are tied into loans
which are classifed as illiquid assets. The liquidity of the banks contracted most
between the years 2006-2007 when the banks were aggressively increasing
their loan portfolios during the country’s preparation for World Cup 2010. The
variation in the ratio from 2007 onwards is ascribed to changes in both loans to
customers and changes in deposit and short term funding. The banks tightened
lending standards later in 2008 in response to the global fnancial crisis such that
credit expansion slipped by 2.6% from R1, 842 billion in 2008 to R1, 794 billion
in 2009 eventually reaching very low levels and zero growth by the end of the
year. The lower rate may also be attributed to a tighter monetary policy stance
and the implementation of the stringent risk based lending criteria by the banks
as a result of the introduction of the National Credit Act of 2007. While bank
loans and advances contracted during 2009, the contraction in credit extension
had both demand-side and supply-side elements (SARB, 2009). On the demand
side, it would appear that households continued to be reluctant to incur more debt
leading to a fall in demand for loans, while on the supply side lending standards
have remained tight and led to the slowing down in the growth of loans and
advances to customers.
Financial stability issues lie at the proftability-liquidity nexus therefore
a decline in liquidity is associated with an increase in proftability, since low
liquidity means larger percentage of assets and total deposits are tied with
loans. Under normal circumstances rapid loan growth tend to result in higher
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
45
returns and higher risks. However, Saunders and Cornett (2006) argue that rapid
growth in assets (loans) than in deposits is indicative of banks using borrowed
funds excessively. This seems to highlight the position of the South African
commercial banks. There has been a growing trend in loans and advances to
customers; however, this increase has not been met by an equal increase in
customer deposits. This may then mean South African commercial banks have
been turning to purchased liquidity in the form of money market instruments to
fund the increase in loans. However, this remains uncertain as the data contained
in Bankscope is not adequate to be able to back up the information with detailed
ratios. Furthermore, a closer inspection of the balance sheets shows that retail
deposits represent only about 25% of total deposits in the South African
commercial banking system, while deposits with less than one year maturity
represent close to 80% of total deposits (IMF, 2008).
It is therefore worth noting that as a result of the above mentioned issues
concerning the composition of the deposits, the South African banking system
faces long-standing structural risks rooted in the sector’s reliance on short-term
wholesale deposits. The IMF has recommended that the SARB explore ways
to reduce the risks associated with the banks’ reliance on short term wholesale
deposits. One of the recommendations is implementing a deposit insurance
system to counter such risks and provide the added beneft of inducing household
saving to migrate from unguaranteed liquid fnancial instruments to competing
bank deposits, thus strengthening the retail base of banks.
5.3 Asset Credit Quality (Credit Performance)
Credit performance is concerned with the examination of the risk associated
with a bank’s asset portfolio. Figure 3 shows a fairly stable trend in the loan
reserve to gross loans ratio between 2005 and 2007. However, 2008-2009 shows
a signifcant deterioration in the credit quality. The ratio registered a decline of
12% from 1.83 to 1.61 during 2005-2006 before slightly increasing from 1.61
to 1.68 in 2007. The slight improvement was due to continued growth in loans
to customers as well as growth in non performing loans which continued on a
downward trend.
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
46
Figure 3: Credit Quality Trend
Data for imperial bank for LRGL for 2009 was missing at the time of data collection
The loan portfolio deteriorated in 2008-2009 as the ratio increased by
8.33% from 1.92 in 2008 to 2.08 in 2009. Credit risk ratios increased during
2008 indicating the deterioration of the quality of the loan portfolio as compared
to 2005. For the period 2008 to 2009 nonperforming loans and advances
increased from 2.8% in July 2008, to 5.5 % in August 2009, as a result of the
crisis and the period was also marked by an increase in credit losses in line with
the tougher market conditions. This is because the banks were more exposed to
increased credit risk as risky loans given during the 2005-2006 period began to
go bad and the banks reported higher charge off or additional provision for loan
losses. Hence South Africa’s banks are facing increased credit risk, especially in
their home loan portfolios, in the face of record household indebtedness and a
mounting debt service burden.
Overall, although risk appetites were adjusted in line with challenging
economic conditions and caution was exercised with regard to lending, the
operating environment continued to be under pressure during 2009 as evidenced
by increased credit impairments and resultant lower proftability levels. This
resulted in the quality of the loan portfolio sharply deteriorating during 2008-
2009.
Hypothesis Testing
To examine whether the difference in performance of the banks in 2005-2006
2005 2006 2007 2009 2008
2.50
2.00
1.50
1.00
0.50
0.00
Loan Reserve-Gross loans
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
47
is statistically different from that of 2008-2009 a student’s t- test is employed to
test the hypothesis that the means of the two periods are the same on the seven
variables. The table below provides a summary of a student’s t- test results for
the two periods under review.
Table 2: Student’s T- Test Results
Mean
2005-2006
Mean
2008-2009
P value Decision
Proftability ROA 1.395 0.986 0.006 Reject
ROE 24.149 16.99 0.009 Reject
C/ I 58.125 48.842 0.0009 Reject
Liquidity NLTA 74.04 73.465 0.666 Accept
NLDST 91.686 82.319 0.391 Accept
LADST 16.746 14.226 0.207 Accept
Credit quality LRGR 1.719 1.997 0.542 Accept
With regard to proftability, ROA and ROE shows banks performed better
in the period 2005-2006 compared to 2008-2009. As shown in Table 2, the mean
for ROA was 1.395 for 2005-2006 compared to the 0.986 for 2008-2009. ROE
shows a similar trend with the mean for 2005-2006 being 24.149 compared
to 16.99 for 2008-2009. This indicates that the banks signifcantly progressed
in proftability during 2005-2006. The P- values for ROA and ROE are 0.006
and 0.009 respectively, therefore the differences between the performances for
the two periods are statistically signifcant as the P-values are below 0.05 and
therefore the null hypothesis has to be rejected leading to the conclusion that
proftability deteriorated during 2008-2009. The C/I ratio shows a similar trend,
the difference in the C/I means is statistically signifcant at 95% confdence
level as the P-value is 0.0009 and therefore less than 0.05 the null hypothesis is
therefore rejected. However, in terms of C/I ratio an improved bank performance
is highlighted for the period 2008-2009 as opposed to ROA and ROE which
showed better performance for 2005-2006.
Liquidity levels have been falling as a result of the fnancial crisis. However,
the null hypothesis of equality of means for the two different time periods
cannot be rejected for NLTA and NLDST as the P-values are 0.666 and 0.391
respectively. This implies that statistically, there is no signifcant difference
between the liquidity performance of the banks in the two periods in terms of
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
48
NLTA and NLDST. The means for the LADST shows a different trend with the
mean for 2005-2006 being 16.746 while the one for 2008-2009 is 14.226. This
indicates that banks have been more liquid in 2005-2006 compared to 2008-2009
in terms of LADST. However, with the P-value being 0.207, the differences are
not statistically signifcant. The improved liquidity is a consequence of more
stringent lending procedure and tightening of credit extension procedures in
response to the fnancial crisis.
With respect to credit quality the mean for loan reserve to gross loan
ratio is 1.719 for 2005-2006 and 1.997 for 2008-2009 indicating that the loan
portfolio deteriorated in 2008-2009. However, the difference is not statistically
signifcant as the P-value is 0.542. Therefore, the null hypothesis cannot be
rejected.
From the results of the student t-test, it can be argued that despite the
fnancial turmoil that engulfed the global economy and affected fnancial
institutions around the world, statistically signifcant differences were only
observed in proftability performance of the South African commercial banks.
On the contrary, no signifcant differences were observed between the overall
performances of the commercial banks in South Africa during the two periods
in terms of liquidity and credit quality. This is supported by the null hypothesis
of the equality of the means being accepted on liquidity and credit quality and
rejected on all three proftability ratios. Although the student’s t-test is showing
a mixture of results on the overall performance, the overall results are consistent
with recent literature (See Baxter, 2008; Mminele, 2009) and shows that South
Africa side-stepped the worst of the crisis and has been resilient to the global
fnancial crisis. This is mainly because South African commercial banks had no
direct exposure to the sub-prime mortgage market, while the banks’ international
franchises had very limited exposure (SARB, 2009).
6. Summary and Conclusion
This paper measured the performance of South Africa’s commercial
banking sector over the period 2005-2009. The results indicate that the overall
bank performance in terms of proftability, liquidity, and credit quality has been
improving since 2005 up to and including 2007. Banks increased the size of
their loan portfolios during the period as the country prepared for World Cup
2010. Although the banks aggressively increased their loan portfolios, sound and
effective credit risk management policies have been in place so that the lending
behaviour could still be contained. This is refected in the downward trend in
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
49
nonperforming loans. However, bank performance deteriorated during 2008-
2009 as the banks’ operating environment deteriorated due to the global fnancial
crisis and a slowing economy. The analysis has also uncovered that the illiquidity
levels in the South African commercial banks has reached extreme levels. This
is exacerbated by the banks’ dependence on wholesale markets and the fact that
deposits with less than one year maturity represent close to 80% of total deposits.
Despite these alarming features, South African banks have managed to continue
with their normal day to day business during the global fnancial crisis. South
African banks’ low leverage, high proftability, and limited exposure to foreign
assets and funding allowed them to remain liquid and well-capitalized; obviating
any need for extraordinary liquidity or state support.
We also found signifcant differences in proftability performance
for the period 2005-2006 and the period 2008-2009. The results indicate that
proftability deteriorated during the later period. There might be several reasons
for the signifcant deterioration in proftability. One of the reasons could be
increasing bank operating costs and reduced incomes amid the global fnancial
crisis. Furthermore in these recessionary times, when corporate and private
clients fnd it hard to service their debts, the level of the provision for loan losses
and bad debts increased. In contrast, no statistically signifcant differences were
observed between bank performance during the two periods in terms of liquidity
and credit quality. The comparable performance results in terms of liquidity
and credit quality between these two periods may be as a result of the fact that
South Africa entered the downturn with a sound macro/fscal position, enabling
aggressive counter-cyclical fscal and monetary responses. Notwithstanding the
turmoil experienced in international fnancial markets and the domestic cyclical
economic developments during 2008-2009, the South African banking system
remained stable; banks were adequately capitalised and proftable. South African
banks remained in a sound position weathering the global fnancial storm, as they
benefted from limited exposure to foreign currency debt and the fact that assets
at the epicentre of the crisis were minimal in South Africa (SARB, 2008).
Notes:
1http://web.worldbank.org/WBSITE/EXTERNAL/COUNTRIES/
AFRICAEXT/SOUTHAFRICAEXTN
2.
Investec Bank has been omitted from the sample as it is an investment bank
rather than a commercial bank. Capitec Bank, Teba Bank and African bank
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
50
have been excluded on the basis that they are micro fnancing institutions
with different profles from the fve banks in the sample.
3. Adjusted liabilities’ refers to total liabilities reduced by (1) funding received
from head offce or from other branches within the same group; and (2)
amounts owing by banks, branches and mutual banks in South Africa.
References
Ahmed, M., B. (2009). Measuring the Performance of Islamic Banks by
Adapting Conventional Ratios German University in Cairo Faculty of
Management Technology Working Paper No. 16 pp 1-26.
Baxter, R. (2008). The global economic crisis and its impact on South Africa
and the country’s mining industry South African Reserve bank conference
papers [Online] Available at:http://www.reservebank.co.za/internet/publication.nsf/WCEV/498F27F900B18
F27422576010035EE92/?opendocument [Accessed 14 June 2010]
Berger, A.N. and Humphrey, D.B. (1997). Effciency of fnancial institutions:
international survey and directions for future research. European Journal of
Operational Research, Vol 98, pp.175-212.
Brand, A (2009). Lagos is not Lehman‘– The Differentiated Impact of the
Global Financial Crisis on Emerging Markets and Developing Countries.
University of Economics, Prague, Publishing house 1st Edition Volume 2
pp. 73-103.
BUREU VAN DIJK (2009). Bankscope Database.
Calem, P. and Carlino, G. (1991) The Concentration/Conduct Relationship in
Bank
Deposit Markets, Review of Economics and Statistics 73 pp. 268-76.
Casu, B., Molyneux, P. And Girardone, C. (2006). Introduction to Banking,
Prentice Hall/ Financial Times, London.
Cronje, T. (2007). Assessing the relative effciency management of South
African banks. Management Dynamics 16 pp 11-23.
Dietrich, J. (1996). Financial Services and Financial Institutions: Value
Creation in Theory and Practice Prentice Hall.
Drake, L. (2001). Effciency and productivity change in UK banking. Applied
Financial Economics 11, pp 557-571.
Gilbert, E., Calitz, E., Plessis, S., (2009). Prudential regulation, its international
background and the performance of the banks a critical review of the South
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
51
African environment since 1970. Economic History of Developing Regions
24 (2), pp 43 — 81.
Greenberg, B. J., & Simbanegavi, W. (2009). Testing for completion in South
Africa Banking Sector Unpublished Manuscript.
Halkos, G. and Salamouris, D. (2004). Effciency measurement of the Greek
commercial banks with the use of fnancial ratios: a data envelope analysis
approach. Management Accounting Research 15 (2), pp 201- 224.
Hempel, G., Simonson, D., and Coleman, A. (1994). Bank Management: Text
and Cases. 4th Edition, John Wiley & Sons, Inc.
IMF (2008). South Africa: Financial System Stability Assessment. IMF
Country Report No. 08/349 [Online] Available onhttp://www.imf.org/
external/pubs/ft/scr/2008/cr08349.pdf [Accessed 19 July 2010]
Kirkpatrick, C., Murinde, V. and Tefula, M. (2007).The measurement and
determinants of x-ineffciency in commercial banks in Sub-Saharan Africa.
European Journal of Finance 14 (7) 2, pp 625-639.
Kiyota, H. (2009).Effciency of Commercial Banks in Sub-Saharan Africa: A
Comparative Analysis of Domestic and Foreign Banks. A paper prepared for
the CSAE conference 2009 on “Economic Development in Africa” held at
the University of Oxford.
KPMG (1998). Bank margins and related proftability in South Africa A study.
[Online], Available on:http://www.scribd.com/doc/14752909/Banking-Ass-
KPMG-Study-Bank-Margins [Accessed 01 August 2010].
Lacewell, S., K. (2003). Do Effcient Institutions Score Well Using Ratio
Analysis? An Examination Of Commercial Banks in The 1990s. Journal of
Commercial Banking and Finance 2, pp 17-33.
Mboweni, T., .T (2004). The South African banking sector - an overview of the
past 10 years. Address by Governor of the South African Reserve Bank, at
the Year-end media cocktail function, Johannesburg, 14 December 2004.
Mminele A.D (2009). Recent economic developments in South Africa:
Citigroup Global Issues Seminar Istanbul 4 October 2009 [Online]
Available onhttp://www.bis.org/review/r091008d.pdf [Accessed 10 August
2010].
Murinde, V. (2009). Global Financial Crisis: Implications for Africa’s Financial
System, paper prepared for the European Development Report (ERD) 2009
Conference on “Financial markets adverse shocks and policy responses in
fragile countries”, 21-23 May, Accra, Ghana.
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
52
Napier, M. (2006). Provision for Financial Services in South Africa in
Liberalisation and universal access to basic services: telecommunications,
Water and Sanitation, Financial Services, and Electricity OECD policy
studies and The World Bank Published by: OECD Publishing.
Ncube, M. (2009). Effciency of the Banking Sector in South Africa, African
Economic Conference 2009 Fostering Development in an Era of Financial
and Economic Crises, Addis Ababa.
Oberholzer, M. and Van der Westhuizen, G. (2004). An Empirical Study
on Measuring Effciency and Proftability of Bank Regions Meditari
Accountancy Research 12 (1), pp 165–178.
O’Donnell, C.J. and van der Westhuizen, G. (2002). Regional comparisons
of banking performance in South Africa. The South African Journal of
Economics 70 (3), pp 485-518.
Okeahalam, C., C. (2006). Production costs and effciency in the South African
banking sector: a stochastic analysis. International Review of Applied
Economics 20, pp 103–23.
Ramanathan, R. (2007). Performance of Banks in Countries of the Gulf
Cooperation Council. International Journal of Productivity and
Performance Management 56 (2), pp 137-154.
Ramcharan, R. (2009). Sound Policies Shield South Africa from Worst of
Recession. IMF African Department IMF survey magazines: countries and
regions survey.
Rossouw, J. (2009). South African Reserve Bank History, functions and
institutional structure SARB [Online] Available onhttp://www.reservebank.
co.za [Accessed 08 June 2010].
Samad, A. (2004). Bahrain Commercial Bank’s Performance during 1994-2001.
Credit and Financial Management Review 10(1) pp 33-40.
Saunders, A and Cornett, M., M. (2006). Financial institutions management: a
risk management approach 5
th
Edition McGraw-Hill.
Seiford, L., M and Zhu, J (1999). Proftability and Marketability of the Top 55
U.S. Commercial Banks Management Science 45 (9), pp 1270-1288.
Sinkey, J. (2002). Commercial Bank Financial Management: In the Financial-
Services Industry 6
th
Edition Prentice Hall, Upper Saddle River, NJ.
South African Reserve Bank (2008). Financial Stability Review, Pretoria: South
African Reserve Bank. [Online] available onhttp://www.reservebank.co.za/
[accessed 15 June 2010].
©2010 The Author (s)
Journal compilation ©2010 African Centre for Economics and Finance
53
South African Reserve Bank Supervision Department Annual Report 2005,
2006; 2007; 2008; 2009 Pretoria [Online] available onhttp://www.
reservebank.co.za/ [accessed 15 June 2010].
Tarawneh, M. (2006). A Comparison of Financial Performance in the Banking
Sector: Some Evidence from Omani Commercial Banks”. International
Research Journal of Finance and Economics 3, pp 103-112.
Thygerson, K., J. (1995). Management of fnancial Institutions, HarperCollins
College Publishers.
Van der Walt S.J. (1998). “The Infuence of Socio-Political Developments on
Banks in South Africa, 1992 to 1997”, Occasional Paper No.13, South
African Reserve
Bank, Pretoria, RSA
Webb, R.M. (2003). Levels of effciency in UK retail banks: a DEA window
analysis. International Journal of the Economics of Business Vol 10 (3), pp
305-322.
World Bank South Africa country Brief [Online] Available onhttp://web.
worldbank.org/WBSITE/EXTERNAL/COUNTRIES/AFRICAEXT/
SOUTHAFRICAEXTN [Accessed 24 July 2010].
Yeh, Q., J. (1996). The Application of Data Envelopment Analysis in
Conjunction with Financial Ratios for Bank Performance Evaluation.
Journal of the Operational Research Society 47, pp 980-988.
doc_422527917.pdf