Efficiency Analysis of Private Commercial Banks in
(First Draft)
By Muluneh Ayalew National Bank of Ethiopia
I. INTRODUCTION
The importance of restructuring and privatizing the financial sector has received renewed attention in the ongoing effort of revitalizing developing economies (World Bank, 1996; Sachs, 1997; Scholtens, 2000). A growing research literature has underscored the importance of banking and the financial sector to economic growth. Several studies found positive relationship between financial sector development and levels of income and growth [Levine, 1997; Khan and Senhadji, 2000].
It is widely believed that efficiency in the banking sector is as crucial for economic growth as it has a direct impact on the productivity of all the other sectors in the economy.
Stable growth, in the context of developing economies requires that the economy be put to a path of higher savings and further ensuring that these savings are chanelised into productive investment. In this scheme of growth, the banking system has a dual role to play. It is a mobilizer of savings as well as an allocator of credit for production and investment. Effectiveness of the banking sector’s contribution to the economic growth and development is broadly determined by its efficiency in the allocation of the mobilized savings amongst competing projects (Misra 2003).
Inclusion of inefficiencies into economic analysis is attractive for several reasons. First, it helps to identify which producers are inefficient and to what extent. Once the inefficient firms are identified policies will be designed to promote efficiency and can be made more effective by directing the necessary help to those who are in the greatest need of assistance. Second, after recognizing the presence of inefficiency; it is natural to examine factors responsible for inefficiency, i.e., identification of determinants of inefficiency. When some explanatory factors are found, programs can be designed and support can be directed to the needy producers to achieve maximum effectiveness (Kumbhakar and Sarkar 2004)
In this paper we explore the extent of concentration of the commercial banking market in Ethiopia using the Herfindahl Index (HI) and the cost efficiency of private commercial banks operating in Ethiopia using stochastic frontier analysis (SFA). By employing the SFA method we will see the impacts of branch expansion, asset size, age, and capital size on the cost efficiency of private commercial banks operating in Ethiopia.
It is hypothesized that concentration in the commercial banking industry in Ethiopia is very high but declining from time to time. The concentration is uniquely significant in total assets and demand deposits and is comparatively lower in loan disbursement and collection. It is also hypothesized that branch expansion and asset size have a negative impact on the cost efficiency of the banks while age and capital size a positive impact.
Though the importance of efficient use of resources has long been recognized; the mainstream neoclassical paradigm in economics assumes that decision-making units in an economy always operate efficiently. Nevertheless, these decision makers are not always efficient. Two firms that are identical in all aspects never produce the same output, and costs and profit are not the same. This output, cost and profit variation can be clarified in terms of technical and allocative inefficiencies, and some other unforseen exogenous shocks. A firm is said to be technically inefficient if it fails to produce the maximum possible output from a given set of inputs. And on the other hand, a cost minimizing or profit-maximizing firm is allocatively inefficient if it fails to allocate the inputs optimally, given the prices of the inputs and outputs. Both inefficiencies are costly in the sense that increase (decrease) cost (profit) arises due to these inefficiencies. Costs of these inefficiencies are also reflected in lower productivity of inputs. That means productivity growth will be lower in the presence of any one, or both, of these inefficiencies (Kumbhakar and Sarkar 2004).
Looking at these inefficiencies for economic analysis is attractive for several reasons. First, it helps to identify inefficient firms and, the extent of their inefficiency. By identifying them policies designed to promote efficiency can be made more effective by directing the necessary help to those who are in the greatest need of assistance. Second, after identifying the presence of inefficiency, factors responsible for the inefficiency would be examined, i.e., determinants of inefficiency would be identified. Once some explanatory factors are found, a program can be designed and support be directed to the needy producers to achieve maximum effectiveness (Kumbhakar and Sarkar 2004).
Many economists have exerted their effort in searching for optimal decisions resulting in efficient allocation of resources. But the problem is how to measure the efficiency. Practically it is impossible to know the frontier of a fully efficient firm; however, it is estimated from observations on a sample of firms in that industry (Coelli, Rao and Battese 1998). For many years, substantial research effort has gone into measuring the efficiency of financial institutions, particularly commercial banks. The research focused mainly on estimating an efficient frontier and measuring the average differences between observed banks and banks on the frontier (Berger and Mester 1997).
Many studies have discovered large inefficiencies, on the order of 20 percent or more of total banking industry costs, and about half of the industry’s potential profits Results of the efficiency estimates often vary substantially across studies according to the data source, as well as the efficiency concepts and measurement methods used. Berger and Humphrey (1997) documented 130 studies of financial institutions’ efficiency, using data from 21 countries, from multiple time periods, and from various types of institutions including banks, bank branches, savings and loans, credit unions, and insurance companies. These variations in the data sets from which efficiencies are measured make it virtually impossible to determine how important the various efficiency concepts, measurement techniques and correlates used are to the outcomes of this studies (Berger and Mester 1997).
Berger and Mester (1997) employed three distinct efficiency concepts, using a number of different measurement methods, and applying comprehensive set of potential efficiency correlates to a single data set. The efficiency concepts employed were -standard profit efficiency, alternative profit efficiency and cost efficiency to estimate the efficiency of 6,000 U.S. commercial banks over six year period of 1990-95.
Standard profit efficiency measures how close a bank is to producing the maximum possible profit given a particular level of input prices and output prices (and other related variables). Unlike the cost function, the standard profit function specifies variable profits in place of variable costs and takes variable output prices as given, rather than holding all output quantities statistically fixed at their observed, possibly inefficient, levels. That is, the profit dependent variable allows for consideration of revenues that can be earned by varying outputs as well as inputs. Output prices are taken as exogenous, allowing for inefficiencies in the choice of outputs when responding to these prices or to any other arguments of the profit function (Berger and Mester 1997).
Berger and Mester (1997) specified the standard profit function, in log form, as:
ln( π+θ) =f(w, vzp) +ln u+ln ∈π(1)
,,
πWhere πis the variable profit of the firm, which includes all the interest and non-interest income of earned on the variable outputs minus variable costs, C, used in the cost function; θis a constant added to every firm’s profit so that the natural log is taken of a positive number; p is the vector of prices of the variable outputs; ln ∈πrepresents the
random error; and ln uπrepresents the inefficiency that reduces profits
The other interesting recent development in efficiency analysis is the concept of alternative profit efficiency, which may be helpful when some of the assumptions underlying cost and standard profit efficiency are not met. Efficiency here is measured by how close a bank comes to earning maximum profits given its output levels rather than its output prices. The alternative profit function employs the same dependent variable as the standard profit function and the same exogenous variables as the cost function. Thus, in lieu of counting deviations from optimal output as inefficiency, as in the standard profit function, variable output is held constant as in the cost function while output prices are free to vary and affect profits (Berger and Mester 1997).
Berger and Mester (1997) specified the alternative profit function, in log form, as:
ln( π+θ) =f(w, y, vz) +ln u+ln ∈aπ(2)
,
aπ
Cost efficiency gives a measure of how close a bank’s cost is to what a best practice bank’s cost would be for producing the same output bundle under same conditions. It is derived from a cost function in which variable costs depend on the prices of variable inputs, the quantities of variable outputs and any fixed inputs or outputs, environmental factors, and random error, as well as efficiency (Berger and Mester 1997). Such a cost function can be written as:
C=vzwyC,φψ) , (3)
(,,, ,
cc
where c measures variable costs, w is the vector of prices of variable inputs, y is the vector of quantities of variable outputs, z indicates the quantities of any fixed netputs (inputs or outputs), which are included to account for the effects of these netputs on variable costs owing to substitutability or complementarily with variable netputs, v is a set of environmental or market variables that may affect performance, ψdenotes an
c
inefficiency factor that may raise costs above the best-practice level, and φdenotes the
c
random error that incorporates measurement error and chances that may temporarily give banks high or low costs. The inefficiency factor ψincorporates both allocative
cinefficiencies from failing to react optimally to relative prices of inputs, w, and technical inefficiencies from employing too much of the inputs to produce y (Berger and Mester 1997).
Cost efficiency is widely defined as the ratio of the total cost to be incurred if the bank was to operate at the most optimal point (i.e., the ideal cost of production) to the actual cost of the bank. The cost efficiency ratio may be thought of as the proportion of costs or resources that are used efficiently. For example, a bank with Cost EFF of 0.70 is 70 percent efficient or equivalently wastes 30 percent of its costs relative to a best-practice firm facing the same conditions. Cost efficiency ranges over (O, 1], and equals one for a best-practice firm. Most researches conducted follow this way.
It is also possible to interchange the numerators and denominators and define cost efficiency as the ratio of actual cost of production to the ideal cost. In this case cost efficiency will be the reciprocal of the former coefficient and is equal to or greater than one, i.e., [1,∞). …….. used this way to estimate the cost efficiency of commercial banks in Botswana, Namibia and South Africa. In the same way, Kraft, Hofler and Payne (2002) followed this method to analyze bank efficiency in Croatia.
Empirical measurement of productive efficiency was first made by Farrell in 1957. Farrell showed how to define cost efficiency and decompose it into its technical and allocative components. He also provided an empirical application to U.S. agriculture, though he didn’t use econometric techniques (Kumbhakar and Sarkar 2004).
Farrell proposed that the overall efficiency of a firm consists of two components. He termed these two components as technical efficiency (which reflects the ability of a firm to obtain maximal output from a given set of inputs) and allocative (or price) efficiency (which reflects the ability of a firm to use the inputs in optimal proportions, given their respective prices and the production technology). Farrell used the term overall efficiency to mean economic efficiency (Coelli, Rao and Battese 1998).
The most common efficiency estimation techniques are Data Envelopment Analysis (DEA), Free Disposable Hull Analysis (FDH), the Stochastic Frontier Analysis (SFA), the Thick Frontier Approach, and the Distribution-Free Approach. The first two techniques are nonparametric while the latter three are parametric methods (Berger and Mester 1997).
Here we focus on the parametric techniques primarily because they correspond well with the cost efficiency concepts. The nonparametric methods generally ignore prices and can, hence, account only for technical inefficiency in using too many inputs or producing too few outputs. They cannot account for allocative inefficiency in misresponding to relative prices in choosing inputs, nor can they compare firms that tend to specialize in different inputs, because there is no way to compare one input with another without the benefit of relative prices (Berger and Mester 1997).
From the parametric methods we focus on Stochastic Frontier Analysis (SFA). SFA had its origin in two papers, one by Meeusen and van den Broeck (June 1977) and the other by Aigner, Lovell and Schmidt (July 1977) (Kumbhakar and Sarkar 2004). The SFA technique starts with a production technology that is specified as:
y=f(x,..., x;β) ×exp{ v+u} (4)
1 k
Where ydenotes output, x1,..., xare kinputs used to produce y, fis the production
ktechnology (black box) which converts inputs to output, and βis a technology parameter vector to be estimated. vis a random noise component, an exogenous shock unknown to the producer. It can be either positive (good luck, for example) or negative. If a producer is unable to produce the maximum possible output, given its input levels and the technology, it is said to be technically inefficient. Such inefficiency might arise due to factors such as, managerial errors arising from inertia and ignorance, poor quality of inputs, etc. Since a technically inefficient firm’s output is always less thanthe maximum possible level determined by the stochastic production frontier (i.e.,
f(x,..., x;β)exp( v)) , given a specific input bundle, a one sided term u(u≤)0 is
1 k
appended to (4) to capture technical inefficiency (Kumbhakar and Sarkar 2004).
In this case inputs are assumed to be given and the objective is to maximize output. Thus, the only inefficiency, if any, is technical. Since data are available only on output and input quantities, estimation of the unobserved inefficiency, u, for each producer from a sample of producers requires some special econometric techniques (Kumbhakar and Sarkar 2004).
The question of resource allocation is not addressed in the above framework because inputs are assumed to be given. In reality, however, input allocation decisions also need to be made. Assuming that the objective of the producer is to minimize cost (of inputs), one can express the technology is terms of the dual cost function viz.,
(
E=wc1,..., w, y;γ/ ) CE(5)
k
Where E is actual cost, c(.) is minimum cost function without any inefficiency, w=(w,..., w) are prices of inputs x1,..., xk, yis output, and γis the technology
1 k
parameter vector (related to βin (5)). CE is the overall cost efficiency. Since actual cost is increased due to technical and allocative inefficiencies, CE ≤1 (Kumbhakar and Sarkar 2004).
Allocative inefficiency arises when the producer fails to use inputs in such a way that the cost is minimized. That means, some inputs are overused and some are underused. Such misallocation leads to an increase in costs. Similarly, compared to another producer who is technically efficient, the presence of technical inefficiency means that an inefficient producer has to use more of every input (which is going to increase cost) to produce a given level of output. This increase in cost due to technical and allocative inefficiencies is captured by the CE term. The reciprocal of CE can be used to measure the percent by which actual cost exceeds the minimum possible cost. The problem here is to i) estimate the overall cost efficiency (CE) and ii) then decompose it into technical and allocative efficiencies. Farrell (1957) showed that the overall cost efficiency (CE) is the product of technical and allocative efficiencies. The decomposition problem is mostly addressed when the technology is known. The problem is much harder in practice, because the task is to estimate an unobserved variable (CE) along with the production technology, and then decompose it into two components (see Kumbhakar (1997)).
Even if the cost-function approach is the dual of the production-function approach of modeling inefficiency, there are some advantages of using the cost-function approach. One advantage is that while the cost-function approach can easily handle cases where producers produce multiple outputs, the production function approach to stochastic frontier analysis is done on the assumption of a single output. It would be rather restrictive to assume a single output in modern day settings where a large number of firms produce multiple outputs (Kumbhakar and Sarkar 2004).
On the other hand, the cost-function approach imposes a behavioral assumption on producers, i.e., producers minimize cost, while the production-function approach does not impose any such behavioral assumption explicitly (although implicitly one assumes output maximization, at least in a single out put framework). However, in competitive environments in which input prices (rather than input quantities) are exogenous, and in which output is also demand driven and so can also be considered as exogenous, the cost-function approach may be more appropriate. In addition, the data requirements for the cost-function approach are higher compared to that for the production function approach. While the latter requires data only on output and inputs, the former requires data on total expenditure, outputs, and input prices. In addition, where a multiple-equations framework is used data on inputs or input-cost share are also required (Kumbhakar and Sarkar 2004, Kumbhakar and Lovell 2000). In this paper we use the cost-function approach towards estimating and modeling inefficiency. Accordingly, we now provide the basic economic framework for estimating these models.
As outlined above, the estimation of a simple equation stochastic cost frontier assumes the existence of data on the prices of the inputs employed, the quantities of outputs produced, and the total expenditure made by each of the I producers. In this case, the estimable cost frontier can be expressed as:
ln E=ln( c( y, w;β)exp{ u}) i=2 ,1 ,..., I(6)
iiii
Where E=∑wxis the actual cost incurred by producer i, y=( y,..., y) ≥0 is
ininii1iMi
n
the vector of outputs produced by producer i, w=(w,..., w) >0 is the vector of input
i1iNi
prices faced by the producer, c( y, w;β) is the cost frontier common to all producers, β
ii
is the vector of technology parameters to be estimated, and u=ln CIcaptures the
i
percentage increase in cost due to inefficiency. Since actual cost is bounded below by the minimum cost c( y, w;β) , the random variable uis non-negative. Higher the value of
iii
uhigher is the cost inefficiency of the producer. Note that in this formulation the input
i
vector xused by the producer ineed not be observed. If this is indeed the case, then
icost inefficiency cannot be decomposed into cost of technical inefficiency and cost of allocative inefficiency (Kumbhakar and Sarkar 2004).
Given the above formulation, the cost efficiency ( CE) of a producer I can be expressed as:
CE=c( yi, wi;β) =exp{ −u} (7)
ii
E
i
Equation (7) defines cost efficiency as the ratio of minimum possible cost to actual or observed cost. Since actual cost is greater than or equal to the minimum cost, it follows that the CEis always less than or equal to one and equals one only when the producer is
iefficient, that is its actual cost equals the attainable minimum cost with that output (Kumbhakar and Sarkar 2004).
In equation (6) the cost frontier c( y, w;β) is deterministic because that the entire excess
iiof observed expenditure over minimum possible expenditure is assigned to cost inefficiency. However, sometimes costs may deviate from the minimum possible due to some other factors than inefficiency, i.e., due to random exogenous shocks like weather, strikes, quality of inputs, etc. which are beyond the control of producers. In order to control for such exogenous factors, another random term is added to the cost function, and the model becomes:
ln E=ln c( y, w;β) +u+vi=2 ,1 ,..., I(8)
iiiii
Under this formulation,
producer-specific random part exp{ vi}. We can calculate the producer specific efficiency exactly as before by:
“…������ ��� ����� ����� ���� �� ���� ����� ���� ������ ��� ����… ����� ����� ��� ���� ��� �� ����� ���� ���� ������� ����� ��� ����… ����� ���� ����� ��� ���� �� ����� �������� ����� �� ���� ��� ��� ����� ���� �� ���� ��� ��� �� ����”
������ ���� ������ (in Amharic) Gebre-Hiwot Bayikedagn, 1924
In 1905 the first modern bank in the history of the country, Bank of Abyssinia was established. The bank was owned and managed by the British-owned National Bank of Egypt and was given a banking monopoly for fifty years, including the right to issue notes and coins. However three other banks were established during the next decade. The first 100 percent African owned bank on the continent, Bank of Ethiopia replaced Bank of Abyssinia in 1931. Bank of Ethiopia was also authorized to issue notes and coins and to act as the government’s bank. Unfortunately after few years of operation the bank was closed following the Italian invasion. Several Italian banks opened branches in Ethiopia during the war. The State Bank of Ethiopia was established in 1942 and became operational in 1943 (Gebre-Hiwot 1924, Belai 1987 and Brownbridge and Harvey 1998).
New banking law split the functions of State Bank of Ethiopia in 1963 in to central and commercial banking, respectively, as National Bank of Ethiopia and Commercial Bank of Ethiopia. Both were owned by government. The 1963 law allowed for other commercial banks to operate, including foreign owned ones provided that they were at least 51 percent owned by Ethiopians (Belai 1987 and Brownbridge and Harvey 1998). Following the law many other banks were established.
In 1975 following the fall of the imperial government there was a major change of economic strategy in the banking sector as it was seen in all other economic sectors. The new government aimed to create a socialist, centrally planned economy on the Soviet model. All privately owned banks were nationalized and concentrated into Commercial Bank of Ethiopia. Then the main financial sector reform was to direct the government banks to finance greatly increased public sector (Brownbridge and Harvey 1998).
Even though economic liberalization began before the fall of Mengistu, neither then nor in the statements of the successor government, did financial sector reform appear as a priority. The government was also very determined not to allow foreign banks into Ethiopia, even as minority partners with Ethiopian banks. The commitment for continued ownership of existing financial institutions was extremely strong (Brownbridge and Harvey 1998). However in recent years measures are being taken to privatize the Construction and Business Bank.
The main institutional changes proposed were very much less radical compared to elsewhere in Africa (Brownbridge and Harvey 1998). Among such changes were: �Allowing private sector banks to operate, but only if owned 100 percent by Ethiopians;
�Restructuring the Agricultural and Industrial Development Bank (now
Development Bank of Ethiopia) and Housing and Savings Bank (now
Construction and Business Bank);
�Giving greater autonomy in lending decisions to Commercial Bank of Ethiopia;
Privatization of banks in developing countries improves bank governance, competition, efficiency, performance and fosters stability. In most developing countries, the government (politicians and bureaucrats) is not a benevolent social guardian and then state-owned banks can be used for political and personal gains. Hence, privatizing of banks (though not sufficient) would be good measure to prevent this to happen. Nevertheless, privatization has some potential costs. Such costs may include private banks turn away from underserved markets (e.g., rural sectors), engage in excessive risk lending and hence engender banking crisis and instability, provide insufficient but concentrated lending if the banking sector is concentrated post-privatization, borrowers with informational and contractual difficulties may be rationed out by private banks (Lemma 2005).
3.2 The Financial System in Ethiopia
In Ethiopia the banking system dominates the financial system. At the close of November 2005 the financial system comprises of one central bank (National Bank of Ethiopia), nine commercial banks (of which two are owned by government), one development bank (Development Bank of Ethiopia), 27 micro-finance institutions (MFIs), one pension fund (Social Security Authority) and numerous savings and credit associations (SCAs) (IMF 2005, Birritu 2006).
At the close of June 2005 total assets of the banking sector (central bank, commercial banks and the development bank) were estimated at Birr 116.5 billion (which is more than 120 percent of GDP of the year). The commercial banks in Ethiopia comprise the publicly owned Commercial Bank of Ethiopia (CBE) and Construction and Business Bank (CBB); and seven other privately owned banks viz., Awash International Bank (AIB), Dashen Bank (DB), Bank of Abyssinia (BOA), Wegagen Bank (WB), United Bank (UB), NIB International Bank (NIB) and Cooperative Bank of Oromia1 (CBO) listed in order of their age.
Unlike their number the privately owned commercial banks have a very small size compared to the public banks, especially the CBE. In addition, all the private banks are 100 percent domestically owned. There is a gradual but encouraging entry of private banks in to the system. The lion's share of the banking market still goes to CBE (Lemma 2005). Encouraging measures were taken to enhance the role of the private sector in the
In this paper, however, when we say private commercial banks we mean all but CBO.
financial sector like liberalizing lending interest rates and exchange rates (Addison and Alemayehu 2001).
Competition takes place where two or more providers of services/ goods put forward their products, as substitutes, to buyers in the same market. It would be difficult to enforce collusion (ant-competitive behavior) in a market where there are several suppliers. In addition, when the firms in the market are of similar sizes competition increases as no one firm could dictate the market (Korsah, Nyarko and Tagoe 2001).
The presence of an uncompetitive market structure leads to low and inefficient financial intermediation. Interestingly, there is no one-to-one relationship between concentration and competition. On the one hand, monopolistic or oligopolistic behavior tends to result in higher intermediation costs and diseconomies of management than under a competitive structure; thus, noncompetitive behavior is consistent with the presence of wide interest rate margins and spreads, which tend to deter potential depositors, as well as potential borrowers, and result in low lending ratios. On the other hand, market size may offer the possibility of exploiting economies of scale (from overhead in administrative operations and information gathering), as well as economies of scope (in combining different product lines for instance). What really matters for the net effect on competition is the
2
level of contestability in the market: the threat of potential competition— or lack thereof—can substantially affect competitiveness conditions, regardless of market concentration (Buchs and Mathisen 2005).
Hence, the incidence of competition is one major factor affecting the efficiency of firms
3
in the market. Higher level of concentration in the market enhances the level of profit but reduces competition Alzaidanin (200?). Hence, it would be wise to see in detail the structure of the market before saying something about the efficiency or profitability of the newly emerged private commercial banks in Ethiopia. In the Ethiopian commercial banking sector, Commercial Bank of Ethiopia seems as having a quasi-monopoly power. In this section it is tried to examine the monopoly power of the CBE using Herfindhal Index (HI). The Herfindahl Index is a concentration measure that can be used as a tool to examine the incidence of competition in a given market. It is given by the formula:
HI=10,000 ∑Si2 (11)
Where Sis the market share of the ithbank
i
The value of the HI varies from 0 (where there is a perfectly competitive industry and the square of the share of one firm is very insignificant and close to zero) to 10,000 (where
2
Buchs and Mathisen (2005) classify the basic idea of market contestability in to two: on the one hand, there are several sets of conditions that can yield competitive outcomes, with a competitive outcome possible even in concentrated systems. On the other hand, collusive actions can be sustained even in the presence of many firms.
3
The conventional definition of concentration is the number and size distribution of firms in the market.
there is a pure monopoly and one firm totally controls the market and the share of (and its square) that firm is one). A market with an HI in excess of 1,800 is generally regarded as highly concentrated and adverse market power effects can be presumed (Korsah, Nyarko and Tagoe 2001 and IMF 2002???).
Competition leads to efficiency and then to cost reduction. Once cost is minimized in the banking sector, cost of borrowing would be lower for other sectors and investment would be enhanced. The Herfindahl Index is of our interest because it is simple to calculate and is used widely to measure concentration in the financial sector.
In applying the Herfindahl Index to the Ethiopian banking industry the problem arises in selecting the indicator of the market. Taking only the balance sheet items may result in biased outcome as most of the balance sheet items like total assets, outstanding loans or total deposits of the commercial banks are functions of time. It is obvious that CBE outweighs all commercial banks with a significant difference in all these items. That is because CBE has advantages over the private banks. These advantages can be classified in to two: one is the government (both central and regional) almost entirely banks with CBE and the other is CBE's much older age compared to others. For years CBE was opening branches throughout the country and establishing good relationship with major sections of the society. Therefore, if we take indicators from the balance sheet only CBE's monopoly power would obviously be higher.
So it would be wise to look beyond the balance sheet and incorporate variables that have a lesser correlation with time than the former ones. In searching for such variables the writer tried to look at collection and disbursement of loans over time on quarterly basis and annual profits of the commercial banks. In addition, it is would be sensible to consider the deposit market in two ways. First taking total deposits and second taking only savings and time deposits. The rationale behind is that CBE as a major banker to government mobilizes almost all deposits of both the central and regional governments. It is known that more than 99 percent of government deposits are put in current account (demand deposits). Hence, it wouldn't surprise if CBE's share in demand deposits is higher than its share in savings and time deposits. To examine this situation the writer preferred to calculate the Herfindhal Index based on the markets for the following variables:
Taking into consideration all these indicators the result supports our expectation. The HI of all indicators of the market demonstrate the quasi-monopoly power of CBE in that all the indicators taken have in average an index of well above 1800. Nevertheless the extent of power differs across indicators. The average share of CBE from the total HI of the market during 2004/05 ranged from 98.4 percent in total assets to 73.4 percent in total loan disbursement (See Table 1 and Fig. 1-6).
Table 1. Herfindhal Index of Some Major Indicators of the Ethiopian Commercial Banking Market
| Particulars | 1999/00 | 2000/01 | 2001/02 | ||||||
|---|---|---|---|---|---|---|---|---|---|
| QI | QII QIII | QIV | QI | QII QIII | QIV | QI | QII QIII | QIV | |
| HI of Total Assets | 7560.5 | 7555.4 7466.4 | 7351.9 | 7087.0 | 7019.2 7088.5 | 7204.8 | 7047.6 | 6961.6 6958.0 | 6802.5 |
| O/w % Share of CBE | 99.5 | 99.6 99.5 | 99.5 | 99.4 | 99.3 99.4 | 99.4 | 99.4 | 99.3 99.3 | 99.2 |
| HI of Total Deposits | 7704.6 | 7627.5 7530.1 | 7444.5 | 7293.5 | 7187.4 7007.3 | 7094.7 | 6921.0 | 6780.3 6601.0 | 6542.7 |
| O/w % Share of CBE | 99.6 | 99.6 99.5 | 99.5 | 99.4 | 99.4 99.3 | 99.4 | 99.3 | 99.2 99.0 | 99.0 |
| HI of Saving and Time Dep. | 6418.7 | 6350.9 6149.1 | 6062.3 | 5859.2 | 5805.6 5596.1 | 5583.5 | 5442.7 | 5357.7 5233.1 | 5091.3 |
| O/w % Share of CBE | 98.7 | 98.6 98.4 | 98.4 | 98.1 | 98.1 97.8 | 97.8 | 97.6 | 97.4 97.2 | 96.9 |
| HI of Loans Outstanding | 6751.7 | 6717.6 6559.5 | 6302.5 | 6102.7 | 5941.8 5783.1 | 5759.2 | 5592.8 | 5420.2 5367.1 | 5308.5 |
| O/w % Share of CBE | 98.8 | 98.8 98.8 | 98.6 | 98.4 | 98.3 98.1 | 98.1 | 97.9 | 97.7 97.6 | 97.5 |
| HI of Loan Disbursement | 5936.5 | 5399.5 3862.6 | 3506.0 | 4922.8 | 3680.5 3861.3 | 3369.2 | 4488.9 | 2894.3 2491.1 | 3112.0 |
| O/w % Share of CBE | 97.6 | 96.6 90.7 | 88.1 | 96.3 | 90.0 89.9 | 86.7 | 95.1 | 78.6 74.6 | 86.4 |
| HI of Loan Collection |