An Evaluation of the Performance of Saudi Banks Abdulbari Ali Alnowahi BA (Business Administration) MA (Economics) Submitted in fulfilment of the requirements for the degree of Doctor of Philosophy School of Business Department of Economics and Finance University of Western Sydney August, 2012
Dedication I dedicate this thesis to the memory of my late parents whom I ask Allah to keep among those who enjoy god s blessings in paradise. I also dedicate this work to my loving wife Nagat and to the light of my life, my children: Hossam, Raghad, Eisam and Hesham for their patience, sacrifices, love and continuous support. To my brothers and sisters, with special thanks to my brother Fouzi for his ongoing encouragement. To all my relatives and friends who believed in my ability and who prayed for me. To all of them I bestow this significant achievement with my hope that it will be a source of more happiness. i
Acknowledgements All praise to Allah, most gracious and merciful, for his help, blessing and guidance. This thesis is the result of hard work, determination and commitment over a number of years during which many obstacles had to be overcome and many sacrifices made, by not only me, but by my immediate family. During my research I learned many skills which I will, hopefully, maintain in my professional and personal life. I must acknowledge the professional and personal support, guidance and encouragement provided to me by my supervisor Professor Satya Paul. His interest and dedication made my tasks easier, achievable and a source of pride. I am forever thankful, appreciative and indebted to Professor Satya Paul, as only through his ongoing commitment and compassion did I manage to achieve my study goal. Sincere thanks to my co-supervisor, Dr Sriram Shanker, for his continuous interest and helpful comments and suggestions in developing this thesis. My sincere thanks to the Saudi Government represented by Ministry of Higher Education for granting my scholarship, and my cordial gratitude to the Saudi Arabian Cultural Mission in Australia for their continuing support and encouragement. I must also appreciate the friendly and productive working environment provided by the department of Economics and Finance. My special thanks to Professor John Lodewijks for his support during his work as Head of the School. Finally, I wish to thank Dr Margaret Johnson of The Book Doctor for professionally editing the thesis. ii
Declaration I declare that the material contained in my thesis is original, to the best of my knowledge, except the material that I acknowledge in my presented study. I further declare that this material has never before been submitted by me at this institution or any other in order to attain a degree or qualification. Signature: Date: iii
Table of Contents Declaration... iii Abbreviations... xi Abstract... xiv Chapter 1 Introduction... 1 1.1 Background... 1 1.2 The Objectives... 3 1.3 The Methodology... 4 1.4 The Structure of the Study... 7 Chapter 2 The Structure of the Saudi Economy... 11 2.1 Introduction... 11 2.2 Composition of Saudi Arabian Population, Labour Force and Basic Social Indicators... 12 2.2.1 Composition of Population... 13 2.2.2 The Labour Force in Saudi Arabia... 14 2.2.3 Basic Social Indicators... 16 2.3 The Structure of the Saudi Arabian Aggregate Supply by Economic Sector and Aggregate Demand by Type of Expenditure... 17 2.3.1 Sub-periods, Oil Prices and the Banking Sector... 23 2.3.2 The World Financial crisis and Government Initiatives... 25 2.4 The Structure of Saudi Foreign Trade and its Balance of Payments... 25 2.4.1 Saudi Exports... 25 2.4.2 Saudi Imports... 29 2.4.3 Balance of Payments... 31 2.5 Summary... 32 Chapter 3... 34 The Development of the Banking Sector in Saudi Arabia... 34 iv
3.1 Introduction... 34 3.2 Primary and Secondary Functions of Commercial Banks... 34 3.2.1 Primary Functions... 34 3.2.1.1 Accepting deposits... 34 3.2.1.2 Granting loans... 36 3.2.2 Secondary Functions... 37 3.3 The Establishment of the Saudi Arabian Monetary Agency (SAMA)... 38 3.4 Commercial Banks in Saudi Arabia: Historical Development... 40 3.4.1 The National Commercial Bank... 40 3.4.2 Samba Financial Group (SAMBA)... 41 3.4.3 Saudi Arabia British Bank (SABB)... 41 3.4.4 Banque Saudi Fransi... 42 3.4.5 Saudi Investment Bank (SAIB)... 42 3.4.6 Saudi Hollandi Bank... 43 3.4.7 Riyadh Bank... 43 3.4.8 Arab National Bank... 44 3.4.9 Al Rajhi Bank... 44 3.4.10 Bank Aljazira (BAJ)... 45 3.4.12 Bank Alinma... 46 3.5 Choice of Banks and Bank Variables... 46 3.5.1 Asset Holdings of Saudi banks... 46 3.5.2 Financial Indicators... 52 3.5.2.1 Net Interest Margin (NIM)... 52 3.5.2.2 Income-Cost Ratio (IC)... 53 3.5.2.3 Ratio of non-interest income to total income... 55 3.5.2.4 Return on Equity... 55 3.5.2.5 Return on Assets... 57 3.6 Summary... 59 Chapter 4 Literature Review... 60 4.1 Introduction... 60 4.2 Efficiency and Productivity Analysis of Islamic, Middle Eastern and Arab banks... 60 4.3 Efficiency and Productivity Analysis of GCC Country Banks... 68 v
4.4 Efficiency and Productivity Analysis of Saudi Arabian Banks... 71 Chapter 5 Technical Efficiency of Saudi Banks... 74 5.1 Introduction... 74 5.2 Alternative Approaches to Measuring Efficiency... 74 5.2.1 Parametric Methods... 76 5.2.1.1 Stochastic Frontier Approach (SFA)... 76 5.2.1.2 Thick Frontier Approach (TFA)... 77 5.2.1.3 Distribution Free Approach (DFA)... 77 5.2.2 Non-parametric Methods... 78 5.2.2.1 Data Envelopment Analysis (DEA)... 78 5.2.2.2 Free Disposal Hull (FDH)... 79 5.2.3 Choice between Non-parametric and Parametric Measures... 80 5.2.4 Data Envelopment Analysis (DEA) Approach to Measuring Technical Efficiency... 81 5.2.4.1 The input-oriented DEA measure of technical efficiency... 81 5.2.4.2 The output-oriented DEA measure of technical efficiency... 84 5.2.4.3 Difference between input- and output-oriented measures of technical efficiency: a simple example... 85 5.2.4.4 Choice of DEA orientation... 86 5.2.4.5 Data Envelopment Analysis Program (DEAP Version 2.1)... 87 5.3 Choice of Variables for DEA Model... 87 5.4 Sample Size and Sources of Data... 90 o An Analysis of the Technical Efficiency of Saudi Banks... 93 5.4.1 Summary Statistics for Inputs and Outputs... 94 5.4.2 Results of Technical Efficiency... 95 5.4.3 Decomposition of Technical Efficiency... 96 5.4.4 Results for DEA Technical Efficiency... 98 5.4.5 Growth Rates of Technical Efficiency... 105 5.4.6 Slacks... 107 5.5 Summary... 110 Chapter 6... 112 Cost Efficiency in Saudi Banks... 112 6.1 Introduction... 112 vi
6.2 Cost Efficiency: Concept and Measurement... 112 6.3 Data and Estimates of DEA-based Cost Efficiency of Banks... 115 6.3.1 Data... 115 6.3.2 Estimates of Cost Efficiency... 116 6.3.3 Growth Rates of Cost Efficiency... 123 6.3.4 Cost Minimizing Input Quantities... 124 6.4 Summary... 127 Chapter 7 Productivity Change of Saudi Banks Based on MPI... 129 7.1 Introduction... 129 7.2 The Malmquist Productivity Index (MPI): Decomposition and Measurement... 130 7.3 Empirical Results... 135 7.4 Summary... 146 Chapter 8 Summary and Conclusions... 149 BIBLIOGRAPHY... 155 vii
List of Tables Table 2.1 Saudi population (in thousands)... 13 Table 2.2 Total Saudi population by sex and nationality (in thousands)... 13 Table 2.3 Population of Saudi Arabia by age (in thousands)... 14 Table 2.4 Composition of the labour force in Saudi Arabia (in thousands)... 14 Table 2.5 Composition of the Saudi labour force by age group and sex in 2008 (in thousands).... 15 Table 2.6 Composition of the non-saudi labour force by age group and sex in 2008... 16 Table 2.7 Gross Domestic Product by kind of economic activity at current price in SR million, and annual growth (within brackets) of each sector, 1998 2008.... 18 Table 2.8 Saudi Expenditure on Gross Domestic Product (SR million)... 22 Table 2.9 Sub-periods and some associated economic indicators... 23 Table 2.10 Value of exports for Saudi Arabia, 2002 2010... 26 Table 2.11 Percentage of Saudi non-oil exports components between 2002 and 2010... 27 Table 2.12 Saudi exports by country group, 2005 2008... 28 Table 2.13 Most important ten countries by value of exports (in million SR)... 29 Table 2.14 Top commodity imports by value, 2010... 29 Table 2.15 Comparison of import components, 2002 and 2008... 30 Table 2.16 Top 15 trading countries by value of imports, 2010... 31 Table 2.17 Saudi balance of payments (million SR).... 32 Table 3.1 Total assets in SR million (at actual price) of 10 Saudi banks, 2001 2010... 48 Table 3.2 Average total assets in SR million (at actual price) for Saudi banks, 2001 2010. 49 Table 3.3 Total assets in SR million (at 2001 constant price) of the 10 Saudi banks, 2001 2010... 50 Table 3.4 Average Assets in SR million (at 2001 constant price) for Saudi banks, 2001 2010... 51 Table 3.5 Net Interest Margin for Saudi banks, 2001 2010... 52 Table 3.6 Income-Cost Ratio for Saudi banks, 2001 2010... 54 Table 3.7 Ratio of non-interest income to total income for Saudi banks, 2001 2010... 55 Table 3.8 Return on Equity (ROE) for Saudi banks, 2001 2010... 56 Table 3.9 Return on Assets (ROA) for Saudi banks, 2001 2010... 58 Table 5.1 List of inputs and outputs... 90 Table 5.2 Classification of Saudi banks based on the 2007 asset holdings of banks... 92 viii
Table 5.3 Summary statistics of variables for Saudi banks, 2001 2010... 94 Table 5.4 Contribution of individual Saudi banks to total banking output, 2001 2010... 95 Table 5.5 DEA estimates of efficiency for Saudi banks, 2001 2010... 100 Table 5.6 DEA estimates of nature of returns to scale for Saudi banks, 2001 2010... 100 Table 5.7 DEA estimates of efficiency by category of Saudi banks, 2001 2010... 102 Table 5.8 Mean DEA estimates of technical, pure technical and scale efficiencies... 103 Table 5.9 Average annual growth rates of efficiencies by... 107 Table 5.10 Sources of inefficiency in input and output variables, (2001 2010)... 109 Table 6.1 Saudi banks input prices, inputs and outputs... 116 Table 6.2 DEA estimates of cost efficiency for Saudi banks, 2001 2010... 117 Table 6.3 DEA estimates of cost efficiency by category of Saudi banks, 2001 2010... 119 Table 6.4 Mean DEA estimates of cost, technical and allocative... 120 Table 6.5 Average annual growth rates of efficiencies by sub-group... 124 Table 6.6 Cost minimizing input quantities for Saudi banks, 2001 2010... 126 Table 7.1 Mean MPI estimates of productivity change and its... 137 Table 7.2 MPI estimates of productivity change and its components, 2001 2010... 138 Table 7.3 MPI productivity change by bank category, 2001 2010... 140 Table 7.4 Average annual growth rates of productivity (%)... 145 ix
List of Figures Figure 3.1 Average total assets (at actual price) of Saudi banks, 2001 2010... 49 Figure 3.2 Average total assets (at 2001 constant price) for Saudi banks, 2001 2010... 51 Figure 3.3 Net Interest Margin for Saudi banks, 2001 2010... 53 Figure 3.4 Income-Cost Ratio for Saudi banks, 2001 2010... 54 Figure 3.5 Return on Equity (ROE) for Saudi banks, 2001 2010... 57 Figure 3.6 Return on Assets (ROA) for Saudi banks, 2001 2010... 59 Figure 5.1 Technical efficiency from an input-orientation... 82 Figure 5.2 Technical efficiency from an output orientation... 84 Figure 5.3 Technical efficiency measures and returns to scale... 86 Figure 5.4 Scale efficiency, scale economies and MPSS... 96 Figure 5.5 DEA estimates of technical efficiency by bank category, 2001 2010... 103 Figure 5.6 DEA estimates of pure technical efficiency by... 104 Figure 5.7 DEA estimates of scale efficiency by... 105 Figure 6.1 Cost, technical and allocative efficiencies... 114 Figure 6.2 DEA estimates of cost efficiency by bank category, 2001 2010... 121 Figure 6.3 DEA estimates of technical efficiency by bank category, 2001 2010... 122 Figure 6.4 DEA estimates of allocative efficiency by bank category, 2001 2010... 122 Figure 7.1 Input distance function and input requirement set... 132 Figure 7.2 Decomposition of Malmquist Productivity Index... 134 Figure 7.3 Annual total factor productivity change, 2001 2010... 142 Figure 7.4 Annual technical efficiency change, 2001 2010... 142 Figure 7.5 Annual pure technical efficiency change, 2001 2010... 143 Figure 7.6 Annual scale efficiency change, 2001 2010... 143 Figure 7.7 Annual technological change, 2001 2010... 144 x
Abbreviations AE Allocative Efficiency ANB Arab National Bank ATM Automated Teller Machines BAJ Bank Aljazira BCC Banker, Charnes and Cooper Model BSF Banque Saudi Fransi CCR Charnes, Cooper and Rhodes Model CE Cost Efficiency CRS Constant Returns to Scale DEA Data Envelopment Analysis DEAP Data Envelopment Analysis Program DFA Distribution Free Approach DMU Decision Making Unit DRS Decreasing Returns to Scale FDH Free Disposal Hull FF Fourier-Flexible f.o.b in Table 2.17 FVIS Fair Value through Income Statement G20 The Group of Twenty Finance Ministers and Central Bank Governors GCC Gulf Cooperation Council GDP Gross Domestic Product HSBC British Multinational Banking and Financial Services Company i.i.d Independent and Identically Distributed IC Income-Cost Ratio IRS Increasing Returns to Scale LP Linear Programming MAX Maximum MENA Middle East and North African Countries MIN Minimum ML Maximum Likelihood xi
MPI MPSS NCB NIM NIRS OE OIC OPEC POS PTE PTEC RIBL ROA ROD ROE SABB SAIB SAMA SAMBA SD SE SEC SFA SHB SR TC TE TEC TFA TFP UAE Malmquist Productivity Index/Indices Most Productive Scale Size The National Commercial Bank Net Interest Margin Non-Increasing Returns to Scale Overall Efficiency Organization of Islamic Countries The Organization of the Petroleum Exporting Countries Point of Sale Pure Technical Efficiency Pure Technical Efficiency Change Riyadh Bank Return on Assets Return on Deposits Return on Equity Saudi Arabia British Bank Saudi Investment Bank Saudi Arabian Monetary Agency Samba Financial Group Standard Deviation Scale Efficiency Scale Efficiency Change Stochastic Frontier Approach Saudi Hollandi Bank Saudi Riyal Technological Change Technical Efficiency Technical Efficiency Change Thick Frontier Approach Total Factor Productivity United Arab Emirates xii
USA VRS WHO WTO United States of America Variable Returns to Scale World Health Organization World Trade Organization xiii
Abstract This study investigates the efficiency and productivity of Saudi banking sector consisting of 10 banks (3 large, 4 medium, 3 small) during the period, 2001-2010. It begins with the estimation of technical efficiency based on input-oriented Data Envelopment Analysis (DEA) approach. For a comprehensive analysis, the technical efficiency is decomposed into the product of pure technical efficiency and scale efficiency. This is followed by measuring cost efficiency which is the product of technical and allocative efficiencies. Finally, the Malmquist Productivity Indices (MPI) are computed to examine total factor productivity (TFP) change over the sample period. TFP change is decomposed into the product of technical efficiency change (catch-up) and technological change. The results reveal a technical efficiency score of 85% for Saudi banking sector. The best performance is shown by small banks followed by medium and large banks respectively. The average cost efficiency for the entire Saudi banking sector is 83.5%. The main source of cost inefficiencies is the technical inefficiency. The TFP change showed an increase of 2.1%, coming mainly from the progress of technological change (TC). xiv
Chapter 1 Introduction 1.1 Background There is growing interest among economists in measuring the efficiency of the commercial banks and other financial institutions in developing and developed economies. However, in comparison to the Western countries, studies focusing on the measurement of banking efficiency and productivity in Arabian and Middle Eastern regions are few. This could be for several reasons. First, Middle Eastern countries are highly regulated and outdated. Second, they are dominated by the public sector and do not face any competition. Third, these economies have introduced financial reforms very recently. Last, reliable data on banks are not available for many countries. In the last fifteen years, many Middle Eastern economies have moved towards liberalizing their financial systems. This has encouraged researchers to undertake studies of banking efficiency and productivity in some of the countries. Jreisat and Paul (2010) examined the cost efficiency level for seventeen Jordanian banks from 1996 to 2007 using the annual data of these banks. Saad and El-Moussawi (2009) examined the efficiency levels of forty-three commercial banks in Lebanon during 1992 using Stochastic Frontier Analysis and data envelopment analysis (DEA). Darrat Topuz, and Yousef (2003) examined the performance of eight Kuwaiti banks 1
for the period between 1994 and 1997, and Romanathan (2006) evaluated the efficiency performance of 55 banks operating in Gulf Cooperation Council (GCC) countries. Mostafa (2007 a) measured the relative efficiency of the top 50 GCC banks using cross-section data for 2005 and the DEA approach. Until now, few studies have evaluated the efficiency and productivity changes of the banking sector in Saudi Arabia. Some of those studies have used frontier approaches, mainly the non-parametric DEA approach. Of these, Al-faraj, Alidi and Bu-Bshait (1993) estimated the relative efficiency of 15 bank branches in Saudi Arabia using a DEA approach. A study by Al-faraj, Bu-Bshait, and Al-Muhammad (2006) aimed to measure the technical efficiency of Saudi banks for the year 2002 using DEA, weighing it against world mean efficiency scores. A third study, conducted by Alkhatlan and Malik (2010), examined both technical and scale efficiencies of ten commercial banks operating in Saudi Arabia between 2003 and 2007, using the DEA approach. Akhtar (2010) used the same approach to measure the efficiency scores and productivity of four Saudi banks during the period 2000 2006. Other studies have used other approaches. These include Haque and Sharme (2011), who studied the financial performance of banks in Saudi Arabia using Speacman s rank correlation method. Essayyad and Madani (2003) measured the concentration, efficiency and profitability of Saudi commercial banks over the period 1989 2001 using financial ratios, while Nader (2010) analyzed the performance of six Saudi Banks using the financial ratios during the period 1998 to 2007. 2
1.2 The Objectives This study investigates the levels of efficiency and productivity of 10 major Saudi banks during the period 2001 2010. The objectives of this study are: 1. To measure the technical efficiency of Saudi banks for the period between 2001 and 2010. 2. To investigate the cost efficiency of Saudi banks for the period from 2001 to 2010. 3. To measure the productivity changes of Saudi banks productivity during the study period. 4. To investigate whether there are differences in efficiency between banks of different sizes. There are several reasons why it is important to study the efficiency and productivity of Saudi banks. Firstly, as they are a major component of the financial system, banks and their functions have a strong influence on economic growth and stability. Secondly, efficiency and productivity are of great significance to Saudi banks because they face both domestic and foreign competition from institutions offering a range of financial services, in a region that is one of the fastest growing in the world. Finally, in comparison to other developed nations, empirical literature on the banking industry in Saudi Arabia is limited. This is due to the small size of the Saudi banking industry and the difficulty in finding relative data. 3
1.3 The Methodology There are two broad approaches to measuring efficiencies and productivity change: the non-parametric approach and the parametric approach. There are two nonparametric methods for measuring efficiency: Data Envelopment Analysis (DEA) and the Free Disposal Hull (FDH). The DEA, developed by Charnes, Cooper and Rhodes (1978), is the more frequently used; it is a linear programming technique for constructing external piecewise frontiers. These frontiers are non-parametric in the sense that they are constructed through the envelopment of the decision making units (DMUs) with best practice DMUs which form the non-parametric frontier. DEA does not impose any functional form specification on the production function. The frontier is formed in such a way that no observation point lies beyond it; therefore, the frontier creates an envelopment of all data points. The FDH model, introduced by Deprins, Simar and Tulkens (1984) and developed by Tulkens (1993), is a special case of the DEA model where the points on lines connecting DEA vertices are not included in the frontier: instead the FDH production possibilities set is composed only of the DEA vertices and the free disposal hull points interior to these vertices. Since the FDH frontier is either congruent with or interior to the DEA frontier, FDH will typically generate larger estimates of average efficiency than DEA (Tulkens, 1993). The DEA is the most widely used popular methodology for measuring efficiency and productivity change. The parametric approach is useful because of its ability to allow for random error, and for the opportunity it offers for mathematical manipulation. A suitable functional form must be selected, which attempts to resemble the actual production process as 4
closely as possible (Coelli et al, 2005). The form can be simple or complex, with varying degrees of complexity between the two extremes. There are three parametric methods to estimate the efficiency/ inefficiency of firms: (i) The Stochastic Frontier Approach (SFA), (ii) The Distribution Free Approach (DFA) and (iii) The Thick Frontier Approach (TFA). A brief discussion of each of these is provided in Chapter 5; it is sufficient to point out here that SFA involves the estimation of efficiencies, usually by estimating either a cost function or a profit function; it allows the testing of hypotheses in regard to the efficiency and structure of production technology. It has been widely used in empirical studies of firms and the banking industry. The difficulties involved in the selection of a distribution form for the efficiency term is a disadvantage of this approach. SFA imposes a particular functional form (and associated behavioural assumptions) that presupposes the shape of the frontier. If the functional form is mis-specified, the measurement of efficiency may be confounded with specification errors. In contrast, non-parametric methods do not impose any structure on the frontier; but they do not allow for random error resulting from luck, data problems or other measurement errors. If random error exists, measured efficiency may be confounded with these random deviations from the true efficiency frontier. We may note that it is not possible to determine which of the two major methods dominates the other since the true level of efficiency is unknown. DEA is a close substitute for SFA; it reports the same measures that SFA does. In light of this, DEA has gained popular acceptance and is frequently applied in studies on bank efficiency and productivity. Hence, the DEA approach has been selected in this study to 5
measure efficiency and productivity change. DEA methodology may be either inputoriented or output-oriented. Since we believe that banks have better control over inputs, the input-oriented DEA approach seems more suitable. Empirical results based on DEA may depend on or are likely to be influenced by the choice and number of inputs entering the model. At the moment there is no agreement on the choice of bank inputs and outputs; in fact, the choice of input and output variables for the banking sector is a matter of some controversy. The literature offers three distinct approaches to selecting inputs and outputs: the production approach, the intermediation approach, and the value-added approach. The first views financial institutions as producers who use inputs of labour and capital to generate outputs of deposits and loans. This approach is used, among others, by Sathye` (2001) and Neal (2004). The intermediation approach views financial institutions as intermediaries that convert and transfer financial assets from surplus units to deficit units. Ahmad (2000) views banks as intermediaries and uses two inputs, labour and deposits, and two outputs, total loans and other investments, for measuring efficiency in Jordanian banks during 1990 1996. In another conceptualization of the intermediate approach, Paul and Kourouche (2008) and Kourouche (2008) use interest expenses and non-interest expenses as inputs, and interest income and non-interest income as outputs. In the value-added approach, high value-creating activities such as making loans and taking deposits are classified as outputs, whereas labour, physical capital and purchased funds are classified as inputs (Wheelock & Wilson, 1995). 6
The intermediation approach is quite popular in empirical research, particularly that based on cross-sectional data (Colwell & Davis, 1992; Favero & Papi, 1995). The production approach, though also found in empirical studies, is less favoured when considering banking as it is known to have limitations, mainly due to the exclusion of interest expenses, which are considered a vital part of banking. There are other practical issues or reasoning governing the selection of inputs and outputs. If the aim is to estimate a unit s production efficiency, then a production approach might be appropriate; however, if the interest of the researcher is in examining intermediation efficiency, then an intermediary approach is appropriate. The choice of variables may also depend on the availability of data. This study uses the intermediation approach, in which banks are viewed as intermediaries that employ two inputs, labour (x1) and total deposits (x2) to produce two outputs, total loans (y1) and other investments (y2). 1.4 The Structure of the Study The study is organized into eight chapters. To put the study into perspective, we begin in Chapter 2 with a brief description of the Saudi economy. The Saudi economy is highly dependent on oil income, and fluctuations in oil prices are likely to affect the GDP and the working of financial institutions in addition to having general effects on other sectors, including the labour market. The chapter focuses on providing a discussion of the composition of the Saudi population and the labour force, and on basic social indicators. It also studies the composition of Saudi foreign trade and its balance of payments, as well as the structure of Saudi aggregate supply, 7
emanating from different economic sectors, and the aggregate demand, based on the type of expenditure. Having discussed the background of the economy, we provide a review of the historical development of the Saudi banking sector in Chapter 3. Since the banking sector in Saudi Arabia modernized not so long ago, the chapter outlines the functions of banks and details of central banking institutions such as Saudi Central Bank and the Saudi Arabian Monetary Agency (SAMA), and the commercial banks operating in the economy. Chapter 4 is devoted to a discussion of the existing literature on bank efficiency and productivity. It starts with the studies conducted in Islamic, Middle Eastern, Arab and GCC countries, and then reviews the studies undertaken of Saudi Arabia. The main research of this study is presented in Chapters 5 through 7. Chapter 5 examines the technical efficiency of 10 Saudi banks during the period 2001 2010 using data envelopment analysis (DEA), a non-parametric approach. A discussion of frontier approaches including DEA is presented. For better understanding, the technical efficiency estimates are decomposed into the products of pure technical efficiency and scale efficiency. The input-oriented DEA model is used to estimate technical efficiency of banks using the DEAP program version 2.1 (Coelli, 1996). The analysis reveals that the level of technical efficiency varies across banks and over time. The technical efficiency score of the aggregated banking sector is 85%, which implies that the total inputs can be reduced by 15% without affecting output, 8
to reach full efficiency level. The group of small banks has the highest average technical efficiency score of 97.1%, followed by medium and large banks. The analysis of technical efficiency, presented in Chapter 5, assumes the absence of allocative efficiency. Chapter 6 makes use of input prices, which permits an investigation of cost efficiency and its decomposition into allocative efficiency and technical efficiency. The empirical results for cost efficiency are obtained by running an input-oriented DEA model, using the software package DEAP Version 2.1 (Coelli, 1996). The results reveal that the average cost efficiency for the entire Saudi banking sector is 83.5%, implying that the average Saudi bank could produce the same level of output using only 83.5% of the resources actually employed, if they were producing on the frontier rather than at their current level. The level of allocative efficiency for the banking sector is 0.93%, higher than the technical efficiency (0.90%). Thus the banking sector is not fully efficient in terms of the allocation of available resources. The levels of cost efficiency and its components vary across banks and over the sample period: in some years some of the banks are found to be fully cost-efficient. The group of small banks shows the highest (98%) level of cost efficiency, followed by the medium (94%) and large banks (90%). The analysis in Chapters 5 and 6 is based on the assumption that the efficiency frontier does not shift over the sample period. However, the efficiency frontier can shift, due to technological progress (technological innovations). Technological progress should be distinguished from gains in technical efficiency represented by units moving toward the frontier (the catching-up effect ). Therefore, Chapter 7 is devoted to investigate changes in total factor productivity (TFP) over time, whether 9
due to technological change (TC) or technical efficiency change (TEC), or a combination of both. In this chapter, we estimate the productivity change of banks using the Malmquist Productivity Index (MPI). The MPI provides a measure of total factor productivity (TFP) change, which is decomposed into the product of technical efficiency change (TEC) and technological change (TC). Technical efficiency change is further decomposed into the product of pure technical efficiency change (PTEC) and scale efficiency change (SEC). This decomposition is useful in that it provides information on the sources of productivity change in banks. The banking sector as a whole shows a TFP change of 2.1% during the study period. The improvement in productivity comes more from progress in technological change than from technical efficiency change The annual changes in TFP are seen progressing over the years except for two, 2006 and 2008. These two anomalous years could be linked to the effect of the local financial crisis in 2006 and the international financial crisis in 2008. Chapter 8 summarizes and brings together the conclusions. 10
Chapter 2 The Structure of the Saudi Economy 2.1 Introduction The Kingdom of Saudi Arabia can be considered a small open economy depending on oil as its major source of income. The Kingdom possesses more than 25% of the world s proven petroleum reserves, currently ranks as the largest exporter of petroleum, and plays a leading role in the OPEC cartel (Mehrara & Oskoui, 2006). The objective of this chapter is to provide an overview of the Saudi economy so that the analysis of banking efficiency and productivity may be put in perspective. For instance, the Saudi economy has experienced internal economic crisis, which might have affected the working/ efficiency of the banking/ financial sectors. This chapter uses data from both local official sources and international sources such as the World Bank and International Monetary Fund, to provide a sketch of the Saudi economy. It shows that the Saudi economy s dependence on an expatriate labour market has increased continuously since the 1980s as the Kingdom strives to meet the rapid growth of the country s development. This reliance has recently begun to decrease, with increases in the Saudi population and the adoption of a Saudization policy by the government. This chapter illustrates that the Saudi economy depends, as do many other oil export countries, on oil production and its export as a main source of the kingdom s income. In other words, the oil sector provides the major contribution to the Saudi GDP compared with other sectors of the economy. The fluctuation of oil prices seems to have a significant effect on both the GDP, in terms both of economic 11
activity and Saudi expenditure. Finally, this chapter addresses the Kingdom s terms of trade with other countries, demonstrating the extent to which its exports and imports have expanded since joining the WTO, and noting that most expenditure on trade is related to the import of private cars. The chapter is divided into five sections. Section 2 analyzes the composition of the Saudi population and labour force, and the basic social indicators. Section 3 studies the structure of Saudi aggregate supply emanating from the different economic sectors and the aggregate demand based on the type of expenditure. Section 4 examines the composition of Saudi foreign trade and its balance of payments. Last section summarizes the chapter. 2.2 Composition of Saudi Arabian Population, Labour Force and Basic Social Indicators Table 2.1 presents the Saudi population for the period from 1970 to 2007. The table shows a sharp increase in population from around 5.7 million in 1970 to more than 23 million in 2007. It also shows that the percentage of non-saudis in the Kingdom s population increased from 5.2% in 1970 to 27.4% in 2007. This increase can be attributed to massive development in Saudi Arabia, particularly after the 1970s, due to sharp increases in the oil price, when it became necessary to allow foreigners to enter, work and live in the Kingdom to meet the need for rapid development. 12
Total Population Saudi Non-Saudi 1970 5744.7 5441.7 303.0 % 100 94.7 5.2 Table 2.1 Saudi population (in thousands) 1980 % 1992 % 9604.4 7800.4 1804.0 100 81.2 18.8 16929.3 12304.8 4624.5 100 72.7 27.3 2004 22678.3 16527.3 6150.9 % 100 72.9 27.1 2007 23678.8 17191.4 6487.5 Sources: 1- Central Department of Statistics & Information, Ministry of Economy and Planning, Kingdom of Saudi Arabia, annual report, 1970 2007. 2- IMF: International Financial Statistics, 2004 and 2007. % 100 72.6 27.4 Table 2.2 Total Saudi population by sex and nationality (in thousands) Year Saudi Non-Saudi Total Populatio M % F % Total M % F % Total n 1970 2877.6 52.9 2564.1 47.1 5441.7 221.9 73.2 81.1 26.8 303.0 5744.7 1980 4090.5 52.4 3709.9 47.6 7800.4 1302.1 72.2 501.9 27.8 1804.0 9604.4 1992 6356.7 51.7 5948.2 48.3 12304. 3254.2 70.4 1370.2 29.6 4624.5 16929.3 8 2004 8288.4 50.2 8238.9 49.9 16527. 4269.9 69.4 1880.9 30.6 6150.9 22678.3 3 2007 8669.6 50.4 8521.8 49.6 17191. 4 4479.6 69.1 2007.9 31.0 6487.5 23678.8 Sources: 1- Central Department of Statistics & Information, Ministry of Economy and Planning, Kingdom of Saudi Arabia, annual report, 1970 2007. 2- IMF: International Financial Statistics, 2004 and 2007 Table 2.2 shows the Saudi population according to sex and nationality for the same period, 1970 to 2007. From this table we observe that the ratio of Saudi females to males increased from 47% in 1970 to more than 49.5% in 2007. Males form about 70% of the total non-saudi population. The labour market attracts males in much numbers than females, because most development is taking place in construction and real estate, and males are more suited for jobs in these sectors. 2.2.1 Composition of Population Table 2.3 shows that the population of Saudi Arabia according to the latest census, in 2004, reached approximately 23 million, of which 73%, or around 17 million, are Saudis and the rest expatriates. Females represent 49% of Saudi residents and males 51%. Of the six million non-saudis, males represent around 70%, and 30% are females. The table also shows that the percentage of the population aged less than 20 13
is about 60%. These characteristics of the Saudi population along with the high growth rate of 2.3% (WHO, country profile, 2008) has put pressure on the government to meet the demand for public services such as, education, health, telecommunication and other types of infrastructure. Table 2.3 Population of Saudi Arabia by age (in thousands) Age Group Saudi Non-Saudi Total Males Females Total Males Females Total Males Females Total Less Than 210.9 206.9 417.9 41.8 39.9 81.8 252.8 246.9 499.7 1 1 4 855.0 847.5 1702.6 181.6 175.8 357.3 1036.6 1023.3 2059.9 5 9 1127.2 1112.5 2239.8 207.0 197.9 404.9 1334.3 1310.5 2644.7 10 14 1081.9 1155.7 2237.6 176.6 167.7 344.2 1258.5 1323.4 2581.8 15 19 948.7 938.9 1887.7 150.3 142.2 292.5 1099.0 1081.2 2180.2 20 24 760.1 786.5 1546.7 298.3 159.9 458.2 1058.4 946.4 2004.8 25 29 725.4 701.3 1426.7 626.7 240.2 866.9 1352.1 941.5 2293.6 30 34 569.2 575.1 1144.3 709.7 271.1 980.8 1278.9 846.2 2125.1 35 39 492.5 498.2 990.7 674.7 190.1 864.8 1167.3 688.3 1855.6 40 44 411.9 372.9 784.9 511.9 121.9 633.9 923.9 494.9 1418.8 45 49 313.3 277.5 590.9 338.1 72.8 410.9 651.5 350.3 1001.8 50 54 222.2 199.4 421.6 193.1 43.5 236.6 415.3 242.9 658.2 55 59 146.1 153.2 299.3 84.9 21.6 106.5 230.9 174.8 405.8 60 64 126.6 134.4 260.9 39.6 14.9 54.5 166.1 149.4 315.5 65 69 103.2 101.2 204.5 16.7 7.8 24.5 119.9 109.1 228.9 70 74 77.3 84.9 162.2 9.8 6.4 16.2 87.1 91.3 178.4 75 79 49.6 37.6 87.2 4.2 2.8 6.9 53.8 40.4 94.2 80+ 66.1 55.8 121.9 4.8 4.5 9.3 70.9 60.3 131.1 Total 8287.4 8239.9 16527.3 4269.9 1881.1 6150922 12557.2 10121.0 22678.3 Source: Central Department of Statistics & Information, Ministry of Economy and Planning, Kingdom of Saudi Arabia, Population Census, 2004. 2.2.2 The Labour Force in Saudi Arabia Table 2.4 Composition of the labour force in Saudi Arabia (in thousands) Year Total labour force Saudi (%) Non-Saudi (%) 1970 1666.0 41.6 58.4 1980 2416.9 43.2 56.8 2004 7,336.8 48.3 51.7 2007 8,224.5 49.3 50.7 Sources: 1. Labour Force Survey, Central Department of Statistics & Information in Saudi Arabia, 1970: 2007. 2. Social Indicators of Development, World Bank, 1980: 2004. Table 2.4 shows the composition of the labour force in the Kingdom. The total labour force grew from 1.7 million in 1970 to 8.2 million in 2007. The Saudi share of the labour force increased from 41.6% to almost 50% recently, with a large influx 14
of Saudis into the labour market. This process, in which Saudi citizens share in the total labour force is increasing, is referred to as the Saudization process. Table 2.5 Composition of the Saudi labour force by age group and sex in 2008 (in thousands). Age group Saudi Female Saudi Male Total Saudi F ratio 15 19 3.2 35.0 38.2 9.1 20 24 105.0 400.6 505.6 26.2 25 29 184.5 660.8 845.2 27.9 30 34 141.3 620.8 762.1 22.8 35 39 102.9 531.9 634.8 19.3 40 44 65.3 427.9 493.2 15.3 45 49 32.5 308.9 341.4 10.5 50 54 13.8 199.4 213.2 6.9 55 59 6.1 118.2 124.4 5.2 60 64 1.1 51.9 53.1 2.2 +65 1.2 66.2 67.4 1.8 Source: Saudi labour Force Survey, Central Department of Statistics & Information in Saudi Arabia, 2008. Table 2.5 shows the composition of the Saudi labour force by age group and sex in 2008. This table covers the labour force for people aged from 15 years to 65+ years. It shows a high percentage of workers, both male and female, in the 25 29 age group. This percentage could be due to the high proportion of Saudi youth among the population. Table 2.5 also shows that females make up 13.39% of the Saudi workers in 2008, compared with 10% in 2000, as jobs are created for Saudi women. 15
Table 2.6 Composition of the non-saudi labour force by age group and sex in 2008 (in thousands). Age group Non Saudi Female Non Saudi male Total Non Saudi F ratio 15 19 4.0 19.8 23.9 16.9 20 24 46.6 151.3 197.8 23.5 25 29 111.9 512.9 624.8 17.9 30 34 190.3 811.9 1002.2 18.9 35 39 155.2 747.3 902.5 17.2 40 44 71.2 599.5 670.8 10.6 45 49 31.3 398.9 430.2 7.3 50 54 13.0 239.5 252.5 5.2 55 59 4.7 118.3 123.0 3.8 60 64 1.5 43.3 44.8 3.3 +65 0.5 23.2 23.7 2.3 Source: Non Saudi labour Force Survey, Central Department of Statistics & Information in Saudi Arabia, 2008. Table 2.6 demonstrates the non-saudi labour force in Saudi Arabia. A high number of non- Saudi workers 30 34 years old. Non-Saudi female workers in 2008 make up 11.6% of the total expatriate working force; Most work as maids or in the medical field. This suggests that the Saudi economy attracts males more than females to fill employment gaps. 2.2.3 Basic Social Indicators Since the 1990s, in order to meet the rising demand for public services caused by increases in population, the government has implemented programs in such fields as education and health care. According to the World Bank (Social Indicators Development Reports 1985; 2008), Saudi Arabia has achieved well in social fields; for instance, in education the number of students increased from 2.3 million in 1985 to more than 8.4 million in 2008. In the same period, expenditure on education as percentage of government expenditure increased from 17% to 27.6%. 33% of the population in 1984 was literate; this increased to 87% by 2008. In the health field, life expectancy increased from 52 in 1970 to 75 in 2007. In addition, hospital beds 16
went from a ratio of one bed per 846 people in 1970 to one per 435 in 2007. This and other social indicators show the success of the government s efforts to increase the welfare of society. Saudi citizens, as do those of many other developed nations, enjoy a variety of free benefits such as medical care, education, interest-free loans for housing and other similar advantages. 2.3 The Structure of the Saudi Arabian Aggregate Supply by Economic Sector and Aggregate Demand by Type of Expenditure The Saudi economy is considered the largest among Arab countries. The Kingdom GDP recorded its highest value for decades with Saudi Riyal (SR) 1,753,503 billion ($ 467,600.8 billion) in 2008, with 22% growth, compared to 13% in 2003. Saudi Arabia, like many other Arab states, to a large degree is moving from a mixed economy to a capitalistic economy. This evolution is a product of unique conditions that prevail in Arab states. For many years these states adopted mixed-economy models with heavy government involvement and support. Recently, this has changed as a result of three forces: a sharp decline in oil revenue, the collapse of the socialist economy in the former Soviet Union and its satellite states, and the emergence of the United States as the sole superpower that directly or indirectly shapes political and economic events in the region. In response, a Sheiko-Capitalist model (Ali, 2009) has been adopted. It is similar to the free market model in that it embraces the market mechanism and unrestricted movement of goods and services. The Saudi economy is characterized by a low inflation rate, a low interest rate, free movement of capital in and out of the country, a fixed exchange rate with a stable currency, a developed infrastructure and an expanding private sector. According to the World Bank s Ease 17
of Doing Business Index (2009), Saudi Arabia has improved its position from 67 th to 23 rd position, and currently is considered number one in the Middle East. Table 2.7 shows two figures: the first represents the gross domestic product (GDP) by type of economic activity at current price in millions of Saudi Riyals; the second stands for the annual growth of each sector and is shown within brackets for the period from 1998 to 2008. Table 2.7 Gross Domestic Product by kind of economic activity at current price in SR million, and annual growth (within brackets) of each sector, 1998 2008. Economic Activity A. Industries & other producers except producers of Gov. services 1998 2000 2002 2004 2006 2008 1-Agriculture; Forestry & Fishing 33,901 (1.5) 34,973 (1.5) 36,101 (1.1) 37,187 (2.0) 39,373 (2.9) 41,050 (2.2) 2- Mining & Quarrying 131,865 262,399 236,926 384,468 668,422 1,005,200 (-35.1) (49.5) (2.9) (30.7) (17.1) (37.2) 3- Manufacturing 58,094 68,290 72,975 95,827 123,912 145,263 (-6.8) (8.8) (5.5) (11.1) (11.9) (8.9) 4- Electricity, Gas and Water 7,860 (3.5) 8,515 (4.2) 9,303 (4.2) 10,406 (5.4) 11,664 (5.8) 12,958 (4.3) 5- Construction 40,314 41,724 44,739 51,141 59,139 71,027 (3.3) (5.8) (3.6) (8.5) (7.6) (9.2) 6- Wholesale & Retail trade, Restaurants & Hotels 42,359 (9.6) 47,832 (4.0) 51,735 (3.9) 58,132 (7.9) 67,868 (8.1) 80,649 (9.0) 7- Transport, Storage & Communication 26,811 (2.7) 29,103 (4.3) 31,934 (4.5) 35,667 (7.4) 41,367 (7.6) 49,550 (9.9) 8- Finance, Insurance, Real estate& Business Services 70,172 (-1.4) 76,204 (3.2) 82,072 (4.1) 91,218 (6.3) 104,798 (7.2) 115,906 (4.9) 9- Community, Social & Personal services 20,864 (2.0) 22,176 (3.7) 24,124 (4.6) 26,478 (5.4) 29,203 (4.8) 32,270 (5.3) 10- Imputed Bank services charge 11,523 (6.1) 13,334 (8.05) 14,714 (5.2) 15,950 (4.6) 17,575 (5.0) 18,825 (3.0) B. Producers of Government services 115,918 (-1.5) 119,123 (2.0) 124,486 (0.7) 155,371 (11.0) 196,386 (11.4) 206,456 (3.1) C. Total Except Import Duties 536,635 (-11.9) 697,007 (17.4) 699,680 (3.0) 929,946 (16.7) 1,324,556 (13.0) 1,741,503 (22.1) Import Duties 10,013 9,650 7,386 8,825 11,025 12,000 (10.0) (0.2) (3.6) (9.1) (9.0) (1.7) Gross Domestic Product 546,648 (-11.5) 706,657 (17.1) 707,067 (3.0) 938,771 (16.7) 1,335,581 (12.9) 1,753,503 (22.0) Sources: 1- Central Department of Statistics & Information, Ministry of Economy and Planning, Kingdom of Saudi Arabia, Annual report, 2000: 2008. 2- IMF, International Financial Statistics Yearbook, 1998: 2006, Washington, D.C. 3- World Bank, World Tables, 2008, Washington, D.C. For Agriculture, Forestry and Fishing, it can be seen that the GDP created by this activity increased from SR 33,901 million in 1998 to SR 41,050 million in 2008. In the same period the annual growth rate of the sector shows fluctuation, with a 18
minimum rate at 0.98 in 2003 and the maximum at 2.9 in 2005 and 2006. This rise and fall is likely due to change in oil price and its effect on the annual growth of other sectors. In addition, reductions in government aid as a result of the constraints on public expenditure required by the budget deficit of the 1990s affected annual growth rates. Although Saudi Arabia is self-sufficient in most food products such as wheat, dates, eggs and dairy, it is still the largest importer of food in the Middle East, with imports reaching $17.8 billion in 2006. For Mining and Quarrying (mainly the crude petroleum and natural gas industry), it is obvious that this sector created the highest proportion of the country s GDP with SR 1,005,200 million in 2008 compared with SR 131,865 million in 1998. The annual growth in this sector shows a continuous increase from -35.12 in 1998 to 37.2 in 2008, as a result of the increase in oil prices and oil production of the Kingdom, particularly in 2008 prior to the world financial crisis, when oil price exceeded $150 a barrel and production reached more than 9.5 million barrels a day. These figures show the importance of this sector to the Saudi economy. For Manufacturing, the share of GDP created reveals a continuous increase from SR 58,094 million in 1998 to SR 145,263 million in 2008. The annual growth rate of this sector fluctuated, with a maximum rate of 18.21 in 2003. The number of factories increased from 3163 in 1998 to 4048 in 2008. For Electricity, Gas and Water, Table 2.7 shows that the share of this activity within the GDP grew from SR 7860 million in 1998 to SR 12,958 million in 2008. The annual growth varied from year to year, reaching its maximum of 6.5 in 2007. In 19
spite of this, the share of this sector in the country s GDP is expected to increase in the near future due to privatization. The construction industry belongs to the private sector, and its annual growth rate rapidly increased from 3.3 in 1998 with a GDP of SR 40,314 million to 9.2 in 2008 with a GDP of SR 71,027 million. The sector at this time was at its best with a large number of new government projects such as building new industrial cities, developing new power and water plants, and constructing a new railway network which is supposed to cover Saudi Arabia and connect it with its neighbours. In addition, natural population growth gives this sector importance, as increased demand for housing and office space necessitates the construction of new residential and business units. Wholesale and Retail trade, Restaurants and Hotels has witnessed a continuous growth since the 1970s with the boom in oil prices. The annual growth rate of this sector is highly correlated with the oil sector performance. The highest annual growth rate of this sector was 9.59% in 1998; it fell to 9.0% in 2008. Transport, Storage and Communication is one of the highest growth sectors in Saudi Arabia. The Kingdom has a solid network of different types of roads, many ports and four international airports: Riyadh, Jeddah, Madinah and Dammam. This infrastructure facilitates the distribution of goods and services within the country and eases contact with the rest of the world. Communication has undergone the same kind of progress, with many companies providing all types of telecommunication nationally and internationally. The annual growth within this sector has gradually 20
increased from 2.68% in 1998, contributing SR 26,811 million to the GDP, to 9.9% and SR 49.550 in 2008. The growth of this sector is expected to continue in the coming years specifically because (a) this sector has been recently privatized, (b) there has been an increase in demand for the services offered by this sector and (c) the Saudi economy has been moving towards greater integration with the developed world. Finance, Insurance, Real Estate and Business Services witnessed the same type of progress as Transport, Storage and Communication with its contribution to the GDP growing continuously from SR 70,172 million in 1998 to SR 115,906 million in 2008. Annual growth fluctuated, recording a peak of 7.2 in 2005 2006. The financial sector has witnessed a fast growth in the past few years, when, according to the Saudi Arabian Monetary Agency (SAMA), the number of banks in the Saudi market reached 18. Among these are five foreign banks, whose branches increased from 1353 in 2007 to 1209 in 2003. SAMA notes that the combined assets for all the commercial banks in Saudi Arabia increased from $145.4 billion in 2003 to $286 billion in 2007. After new rules were adopted by the Saudi government, there was a dramatic increase in the number of new insurance companies. In 2007 there were twenty; there was only one in 1985. Community, Social and Personal services shows a slow growth rate between 1998 and 2008 with a contribution to the GDP of SR 20,864 and 32,270 million respectively. The annual growth for this sector increased from 2.04 in 1998 to 5.3 in 2008, and it is expected to continue to rise with the increased demand for community and social services that has become more insistent post-privatization. 21