## 1. Number of Sample Banks by Countries, 1997-2003

## Source details

**Canonical URL:** [1. Number of Sample Banks by Countries, 1997-2003](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp05117.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp05117.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp05117.pdf.json)

---

### Introduction: context and motivations
- Bahrain banking sector scale and structure:
  - Total commercial bank assets in excess of 120 percent of GDP.
  - Offshore banking sector is "an impressive 24 times of Bahrain’s GDP" but offshore banking units (OBUs) do not participate in domestic economic activities outside of providing very limited co-financing for large public-sector-led infrastructure projects.
  - Bahrain Monetary Agency (BMA) consolidated supervision and regulation of the financial sector in 2002.
  - Financial-sector advantages listed: same-day regional transactions, well-developed telecommunications infrastructure, market openness, easy access to GCC markets, a tax-free environment, no exchange restrictions, a well-trained labor force, high standards of living, stable exchange rate, and a solid regulatory environment.
- Recent performance and concerns:
  - Commercial banks generated $172 million in profits (with Islamic banks doubling their returns compared to 2002); the sector more than tripled its annual earnings between 2002 and 2003.
  - Identified sector concerns: high overhead costs, problems with the quality of loan portfolio, high lending margins, a large share of nonperforming loans carried over from the past, vulnerability of new loans to collateral-related imperfections and excessive consumer lending.
- Study objective:
  - Assess performance of Bahrain banks relative to banks in Kuwait, Qatar, U.A.E., and Singapore using Data Envelopment Analysis (DEA) to inform reform agenda.

### Methodology: Data Envelopment Analysis (DEA) specification
- DEA characterization and key relationships:
  - Nonparametric linear-programming technique that builds an efficiency frontier from best-performing Decision-Making Units (DMUs) and measures relative performance of other DMUs.
  - Farrell (1957) proposed piece-wise linear frontier; Charnes, Cooper, and Rhodes (1978) implemented via linear programming and coined "DEA". Banker, Charnes, and Cooper (1984) extended to variable returns to scale.
  - Efficiency score θi: θi < 1 implies inefficient (inside frontier); θi = 1 implies efficient (on frontier).
  - Overall technical efficiency (OTE) decomposed as: OTEi = PTEi ⋅ SEi.
  - PTE (pure technical efficiency) calculated relative to variable returns to scale frontier; SE (scale efficiency) captures deviations due to increasing/decreasing returns to scale.
- Baseline inputs and outputs (three each):
  - Inputs: (i) personnel expenditures (proxy for labor input), (ii) fixed assets (proxy for premises, branch network, equipment), (iii) interest expenditures (proxy for leveraged funds).
  - Outputs: (i) revenues (interest + non-interest income), (ii) net loans (loans net of loan loss provisions), (iii) liquid assets (cash + treasury bill holdings + balances with monetary authorities).
- Data source:
  - BankScope database, with Bahrain updates/additions from the BMA.

### Results of efficiency analysis: baseline findings and diagnostics
- Cross-country summary (1997-2003 unless otherwise noted):
  - Singapore (sample of 16 banks for 1997-2002): average OTE = 0.70, PTE = 0.88, SE = 0.81.
    - Earlier study for 1991-1997 reported OTE = 0.74, PTE = 0.86, SE = 0.87 for 10 banks.
  - Bahrain ahead of the GCC sub-group overall; competition from U.A.E. and Qatar is strong.
  - "Banks in Bahrain could be 43 percent more cost effective (i.e., use less inputs to produce the same amount of output, as their most efficient counterparts)."
  - Corresponding figures for 1997-2002: U.A.E. 47 percent, Qatar 46 percent.
- Scale efficiency (SE) averages (as reported):
  - U.A.E. average SE = 0.87 for 1997-2002.
  - Singapore average SE = 0.81.
  - Bahrain average SE = 0.83.
  - Qatar average SE = 0.84.
  - Kuwait average SE = 0.68.
- Bahrain decomposition (1997-2003):
  - Average SE for Bahrain = 0.83.
  - Average PTE for Bahrain = 0.70.
  - Conclusion: inefficiencies largely caused by pure technical inefficiencies and to a lesser extent by scale inefficiencies.
- Trends and statistical notes:
  - PTE index shows trend improvement across all sample countries, suggesting reductions in the main source of inefficiency (consistent with IT developments, new instruments/products, and policy-specific developments).
  - Table 2 in the source reports counts of banks with OTE = 1 by country and year (numbers and parentheses for OTE > 0.9 preserved in original table).

### Bahrain by bank type (Table 3 summary)
- Conventional Banks:
  - Average OTE: 0.54; PTE: 0.69; SE: 0.80.
  - Maximum OTE/PTE/SE: 1.00/1.00/1.00.
  - Minimum OTE/PTE/SE: 0.33/0.43/0.50.
  - St. Deviation OTE/PTE/SE: 0.19/0.22/0.17.
- Islamic Banks:
  - Average OTE: 0.57; PTE: 0.61; SE: 0.91.
  - Maximum OTE/PTE/SE: 1.00/1.00/1.00.
  - Minimum OTE/PTE/SE: 0.30/0.37/0.57.
  - St. Deviation OTE/PTE/SE: 0.27/0.24/0.14.
- Reported T-test results: 0.11 (OTE), -0.31 (PTE), 0.66 (SE).

### Robustness tests and diagnostics
- Sub-sample robustness:
  - Bahrain and Singapore only, reduced to two inputs (fixed assets, interest expenditures) and two outputs (revenues, liquid assets) for 1997-2002.
  - Result: relative pattern between Singaporean and Bahraini banks preserved; breakdown between PTE and SE preserved.
- Sensitivity to specification:
  - Excluding loans as output and personnel expenditures as input narrows the efficiency gap between Singapore and Bahrain.
  - Implication: possible inefficiencies related to (i) scope of operations/products (loans) and/or (ii) employment resources (use of labor, high wages/benefits).
- Data caveat:
  - Available data do not allow differentiation between price and quantity of inputs (labor), limiting inferences about efficiency of their use to overall value only.

### Conclusions and policy-relevant implications
- Summary conclusions:
  - Bahrain occupies a front-runner position among the sample GCC countries but lags behind Singapore.
  - Banks in Bahrain operate at similar scale efficiency levels as banks in Singapore and competitors in Qatar and U.A.E.
  - Results robust to changes in sample size and model specifications.
- Policy implications and recommended directions:
  - Further analysis needed to pinpoint exact types of inefficiencies, including product-mix and scope inefficiencies.
  - Investigate underlying sources of cross-country performance differences: macroeconomic and prudential environments, size and concentration of sectors, banks’ ownership structure, and other institutional factors.
  - Sustain regional competitiveness by continuous streamlining of regulatory restrictions, entry and exit rules, and collateral-related bureaucratic practices to limit detrimental effects on banking efficiency.
  - Optimal banking-sector architecture would require regional and international integration to guarantee transfers of new skills and technology through competitive pressures and the search for more profits.

*Source: _wp05117 - 1. Number of Sample Banks by Countries, 1997-2003.*

### 1. Number of Sample Banks by Countries, 1997-2003 ...........................................................14

### 1. Number of Sample Banks by Countries, 1997-2003 ...........................................................14

### Listed tables and sections
- 1. Number of Sample Banks by Countries, 1997-2003 ...........................................................14
- 2. Number of Efficient Sample Banks by Country and Year ..................................................15
- 3. Bahrain: Distribution of Efficiency Indexes by Types of Banks .........................................15

### Figures referenced in the unit
- Figure 1. Conventional Indicators of Bank Performance, 1997-2002 ................................................11
- Figure 2. Cross-Country Comparisons of Efficiency Indexes, 1997-2003 .........................................12
- Figure 3. A Robustness Test: Two-Input-Two-Output Case ..............................................................13

*Source: _wp05117 - 1. Number of Sample Banks by Countries, 1997-2003 ...........................................................14*

### References..............................................................................................................

### _wp05117 - References..............................................................................................................

### Introduction: context and motivations
- The banking sector in Bahrain has total commercial bank assets being in excess of 120 percent of GDP.
- The offshore banking sector is "an impressive 24 times of Bahrain’s GDP" but offshore banking units (OBUs) do not participate in domestic economic activities outside of providing very limited co-financing for large public-sector-led infrastructure projects.
- The Bahrain Monetary Agency (BMA) consolidated supervision and regulation of the financial sector in 2002.
- Bahrain's financial-sector advantages listed in the source: same-day regional transactions, well-developed telecommunications infrastructure, market openness, easy access to GCC markets, a tax-free environment, no exchange restrictions, a well-trained labor force, high standards of living, stable exchange rate, and a solid regulatory environment.
- Commercial banks generated $172 million in profits (with Islamic banks doubling their returns compared to 2002); the sector more than tripled its annual earnings between 2002 and 2003.
- Identified sector concerns: high overhead costs, problems with the quality of loan portfolio, high lending margins, a large share of nonperforming loans carried over from the past, vulnerability of new loans to collateral-related imperfections and excessive consumer lending.
- Paper motivation: assess performance of Bahrain banks relative to banks in Kuwait, Qatar, U.A.E., and Singapore using Data Envelopment Analysis (DEA) to inform reform agenda.

### Methodology: Data Envelopment Analysis (DEA) specification
- DEA overview:
  - Nonparametric linear-programming technique that builds an efficiency frontier from best-performing Decision-Making Units (DMUs) and measures relative performance of other DMUs.
  - Farrell (1957) proposed piece-wise linear frontier; Charnes, Cooper, and Rhodes (1978) implemented via linear programming and coined "DEA".
  - Banker, Charnes, and Cooper (1984) extended to variable returns to scale.
- Key formulation relationships preserved from source:
  - Efficiency score θi: θi < 1 implies inefficient (inside frontier); θi = 1 implies efficient (on frontier).
  - Overall technical efficiency (OTE) decomposed as: OTEi = PTEi ⋅ SEi.
  - PTE (pure technical efficiency) calculated relative to variable returns to scale frontier; SE (scale efficiency) captures deviations due to increasing/decreasing returns to scale.
- Inputs and outputs used in baseline specification:
  - Inputs (three): (i) personnel expenditures (proxy for labor input), (ii) fixed assets (proxy for premises, branch network, equipment), (iii) interest expenditures (proxy for leveraged funds).
  - Outputs (three): (i) revenues (interest + non-interest income), (ii) net loans (loans net of loan loss provisions), (iii) liquid assets (cash + treasury bill holdings + balances with monetary authorities).
- Data source: BankScope database, with Bahrain updates/additions from the BMA.

### Results of efficiency analysis: baseline findings and diagnostics
- Cross-country summary (1997-2003 unless otherwise noted):
  - Singapore banks (sample of 16 banks for 1997-2002): average OTE = 0.70, PTE = 0.88, SE = 0.81. (Earlier study for 1991-1997 reported OTE = 0.74, PTE = 0.86, SE = 0.87 for 10 banks.)
  - Bahrain appears ahead of the GCC sub-group overall; competition from U.A.E. and Qatar is strong.
  - Based on average overall indexes for 1997-2003, "banks in Bahrain could be 43 percent more cost effective (i.e., use less inputs to produce the same amount of output, as their most efficient counterparts)."
  - Corresponding figures for 1997-2002: U.A.E. 47 percent, Qatar 46 percent.
- Scale efficiency (SE) comparisons (averages reported in source):
  - U.A.E. average SE = 0.87 for 1997-2002.
  - Singapore average SE = 0.81.
  - Bahrain average SE = 0.83.
  - Qatar average SE = 0.84.
  - Kuwait average SE = 0.68.
- Bahrain decomposition (1997-2003 as reported):
  - Average SE for Bahrain = 0.83.
  - Average PTE for Bahrain = 0.70.
  - Conclusion: inefficiencies largely caused by pure technical inefficiencies and to a lesser extent by scale inefficiencies.
- Trends and statistical notes:
  - PTE index shows trend improvement across all sample countries, suggesting reductions in the main source of inefficiency (consistent with IT developments, new instruments/products, and policy-specific developments).
  - Table 3 summary for Bahrain by bank type (averages preserved exactly as in source):
    - Conventional Banks — Average OTE: 0.54; PTE: 0.69; SE: 0.80; Maximum OTE/PTE/SE: 1.00/1.00/1.00; Minimum OTE/PTE/SE: 0.33/0.43/0.50; St. Deviation OTE/PTE/SE: 0.19/0.22/0.17.
    - Islamic Banks — Average OTE: 0.57; PTE: 0.61; SE: 0.91; Maximum OTE/PTE/SE: 1.00/1.00/1.00; Minimum OTE/PTE/SE: 0.30/0.37/0.57; St. Deviation OTE/PTE/SE: 0.27/0.24/0.14.
    - T-test results reported: 0.11 (OTE), -0.31 (PTE), 0.66 (SE).
  - Table 2: counts of banks with OTE = 1 by country and year are reported in the source (numbers and parentheses for OTE > 0.9 preserved in original table).
- Robustness tests:
  - Sub-sample test: Bahrain and Singapore only, reduced to two inputs (fixed assets, interest expenditures) and two outputs (revenues, liquid assets) for 1997-2002.
  - Findings: relative pattern between Singaporean and Bahraini banks preserved; breakdown between PTE and SE preserved.
  - Excluding loans as output and personnel expenditures as input narrows the efficiency gap between Singapore and Bahrain — implying possible inefficiencies related to (i) scope of operations/products (loans) and/or (ii) employment resources (use of labor, high wages/benefits).
  - Source caveat: available data do not allow differentiation between price and quantity of inputs (labor), limiting inferences about efficiency of their use to overall value only.

### Conclusions and policy-relevant implications
- Summary conclusions:
  - Bahrain occupies a front-runner position among the sample GCC countries but lags behind Singapore.
  - Banks in Bahrain operate at similar scale efficiency levels as banks in Singapore and competitors in Qatar and U.A.E.
  - Results robust to changes in sample size and model specifications.
- Policy implications and recommended directions (as stated in the source):
  - Further analysis needed to pinpoint exact types of inefficiencies, including product-mix and scope inefficiencies.
  - Policymaker-relevant investigations should examine underlying sources of cross-country performance differences: macroeconomic and prudential environments, size and concentration of sectors, banks’ ownership structure, and other institutional factors.
  - Sustaining regional competitiveness requires continuous streamlining of regulatory restrictions, entry and exit rules, and collateral-related bureaucratic practices to limit detrimental effects on banking efficiency.
  - Optimal banking-sector architecture would require regional and international integration to guarantee transfers of new skills and technology through competitive pressures and the search for more profits.

*Source: _wp05117 - References.*

### References

### _wp05117 - References

### Foundational DEA and efficiency measurement literature
- Charnes A, W. W. Cooper, and E. Rhodes, 1978. Measuring the Efficiency of Decision Making Units. European Journal of Operations Research, No. 2, 429-44.  
- Farrell, M. J.,  1957. The Measurement of Profit Efficiency. Journal of the Royal Statistical Society, Series A, CXX, Part 3, 253-290.  
- Banker, R.D., A. Charnes and W.W. Cooper, 1984. Some Models for Estimating Technical and Scale Inefficiencies in Data Envelopment Analysis. Management Science, 30, 1078-1092.  
- Färe, R., S. Grosskopf, and C.A. K. Lovell, 1985. The Measurement of Efficiency of Production, Kluwer Academic Publishing, Boston, MA.  
- Seiford, L. M., 1996. Data Envelopment Analysis: The Evolution of the State of the Art (1978-1995). Journal of Productivity Analysis, Vol. 7, July 1996: 99-137.  
- Cooper, W.W., L. M. Seiford and K. Tone, 2000. Data Envelopment Analysis. Kluwer Academic Publishers, Norwell, MA.  

### Applications to banking and financial institutions
- Berger A. N. and D. B. Humphrey, 1997. Efficiency of Financial Institutions: International Survey and Directions for Future Research. European Journal of Operational Research, Vol. 98, No. 2, 175-212.  
- Bergendahl G., 1998. DEA and Benchmarks - An Application to Nordic Banks, Annals of Operations Research, Vol. 82, 233-249.  
- Rezvanian, R., and S. Mehdian, 2002. An Examination of Cost Structure and Production Performance of Commercial Banks in Singapore, Journal of Banking and Finance, Vol. 26, 79-98.  
- Grigorian, D., and V. Manole, 2002. Determinants of Commercial Bank Performance in Transition: An Application of Data Envelopment Analysis, IMF Working Paper 02/146.  

* _wp05117 - References *

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp05117.pdf_