The effects of bank regulations, competition, and financial reforms on banks' performance

Emerging Markets Review 12 (2011) 1–20

Contents lists available at ScienceDirect

Emerging Markets Review j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / e m r

The effects of bank regulations, competition, and financial reforms on banks' performance☆ Sami Ben Naceur a,⁎, Mohammed Omran b a

ESSEC Tunis, 4 Rue Abou Zakaria Al Hafsi, Tunis 1089, Tunisia Cairo and Alexandria Stock Exchanges, 4 (A) El Sherifein St., Down Town, Postal Code 11513, P.O. Box 358 Mohammed Farid, Cairo, Egypt

b

a r t i c l e

i n f o

Article history: Received 20 June 2009 Received in revised form 10 August 2010 Accepted 31 August 2010 Available online 9 September 2010 JEL classification: E44 G21 L51 Keywords: Bank interest margin Bank profitability Dynamic panel data and MENA region

a b s t r a c t In this paper, we examine the influence of bank regulation, concentration, and financial and institutional development on commercial bank margins and profitability across a broad selection of Middle East and North Africa (MENA) countries. The empirical results suggest that bank-specific characteristics, in particular bank capitalization and credit risk, have a positive and significant impact on banks' net interest margin, cost efficiency, and profitability. Also we find that macroeconomic and financial development indicators have no significant impact on net interest margins, except for inflation. Regulatory and institutional variables seem to have an impact on bank performance. © 2010 Published by Elsevier B.V.

1. Introduction During the late 1980s and the 1990s, several Middle East and North Africa (MENA) countries (Tunisia, Morocco, Egypt, and Jordan, among others) underwent noteworthy financial reforms under the auspice of the International Monetary Fund (IMF). These reforms have significantly affected both the banking system and the domestic stock market.

☆ The views expressed herein are those of the author and should not be attributed to the Cairo and Alexandria Stock Exchanges (CASE), its board of directors, or its management. The paper has received a grant from the seven round of ERF/GDN project. ⁎ Corresponding author. E-mail addresses: [email protected] (S.B. Naceur), [email protected] (M. Omran). 1566-0141/$ – see front matter © 2010 Published by Elsevier B.V. doi:10.1016/j.ememar.2010.08.002

2

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

While several research studies on bank performance have been conducted extensively for US commercial banks and, to lesser extent, for financial institutions in Europe and in large emerging markets (Brazil, China, and others), relatively little is known about bank performance among banks in other regions, in particular, MENA countries. The main contributions of the paper are as follows. Using bank level data from ten countries over the period 1989–2005 (Tunisia, Bahrain, Egypt, Jordan, Kuwait, Lebanon, Morocco, Oman, Saudi Arabia, and United Arab Emirates), our study aims at assessing the impact of financial development, bank regulations, market structure, and institutional factors on bank performance. In addition to the assessment of the above relationship, our study provides an insight into the characteristics and practices of successful commercial banks in terms of performance. In view of the findings, we should be able to draw some policy implications that may be useful for bank management, policymakers, and shareholders in the MENA region. Estimating the source of bank performance using a dynamic system General Method of Moment (GMM) specification, we find that bank specific characteristics, in particular bank capitalization and credit risk, are key determinants. We fail, however, to find any significant relationship between macroeconomic variables and bank performance except for inflation. Also, the results suggest that banks enjoy lower operating costs in a well-developed banking sector environment. Furthermore, the stock market development variable is always positive and significant in all specifications, suggesting that banks operating in a well-developed stock market environment tend to have greater profit opportunities. The regulatory and institutional variables seem to have an impact on bank performance as the results suggest that corruption increases the cost-efficiency and net-interest margins while an improvement in the law and order variable decreases the cost efficiency without affecting performance. These results indicate the need of MENA banks to operate in a more competitive environment, well-developed capital markets, and a better governance environment. The rest of the paper is organized as follows. We document and discuss financial reform efforts in the MENA region in Section 2, concentrating on the banking sector compared with the security market. A literature review of the determinants of bank performance is given in Section 3, distinguishing between single and cross-country studies. In Section 4, we provide a detailed description of data, methodology, and empirical models that includes measurements of our variable of interest. We then report our empirical results and findings in Section 5, while Section 6 concludes the paper and spells out some policy implications.

2. Financial reform in MENA Research on financial development and its relationship with growth is extensive and ongoing. Although the relationship between financial development and growth continues to be debated, there is general agreement that financial repression or government intervention impose on the financial sector restrictions and price distortion that inhibit growth prospects. In addition there is a consensus that macroeconomic stability is critical for the growth of financial services. Thus, countries should adopt appropriate macroeconomic policies, encourage competition within the financial sector, and develop a strong and transparent institutional and legal framework for financial sector activities. In particular, there is need for prudential regulations and supervision, strong creditor rights, and contract enforcement. Therefore, government decision-makers should eliminate financial repression and support financial development as important elements of their policy package to stimulate and sustain economic growth (Creane et al., 2003). In this context, MENA countries have perceived the importance of financial sector reform in allocating investment and enhancing productivity by identifying promising projects and firms, mobilizing savings, encouraging good corporate governance, and enabling the trading, hedging, and diversifying of risk, as well as facilitating the exchange of goods and services. Thus in the late 1990s a number of MENA countries have adopted a financial reform agenda. While the restructuring initiatives in the MENA region are less vibrant than those taking place in Eastern Europe and parts of Asia, nevertheless, several MENA countries are witnessing a new era in privatization, bank regulation, market-orientation, and integration of privately owned banks of different

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

3

organizational structure, resulting in a well-developed, profitable, and efficient banking sector in some countries, particularly the Gulf Cooperation Council (GCC) countries. The economy of the entire MENA region is classified as a bank-based economy as banks are the dominant financial institutions, since they control most financial flows and possess most financial assets. However, economic reforms have directed more bank ownership and activity toward the private sector and have activated the dormant securities market. For decades many governments in the MENA region – apart from the Gulf Countries – adopted financial repression policies that resulted in a nominal interest-rate ceiling that is below the prevailing inflation rate and the rate of currency depreciation. Under repressive regimes, monetary authorities impose high reserve requirements, bank-specific credit ceilings, and selective credit allocation. These measures result in a noncompetitive and segmented financial system. Such polices allowed the authorities to control better the money supply, serving some social goals, such as protecting lenders against usury practices by moderating the free determination of interest rates and keeping interest rates below market rates, which reduces the cost of servicing government debts. As mentioned above, MENA region can be classified as a bank-based economy, therefore, many countries have witnessed a comprehensive financial reform agenda, concentrated on banking reform in the late 1990s. Before this date, both Lebanon and Morocco had a more liberalized financial sector compared with the rest of the region, while other countries tended to have a state-dominated and excessively regulated financial sector, especially Algeria, Libya, and Syria. The underlying argument is that the soundness of the banking system is important not only because it limits economic downturns related to financial panics, but also because it avoids adverse budgetary consequences for governments. Thus prudential regulation is meant to protect the banking system by inducing banks to invest wisely (Murinde and Yaseen, 2004). There were and still ongoing reform in the banking sector in the MENA region. Most of countries in this region went through several bank laws, which contain provisions for disclosure and transparency in the central bank's activities. Central banks are focusing on many policies, including formulating monetary, credit, and banking policy, supervising policy implementation, managing the national gold and foreign exchange reserves, regulate the banking system, managing public debt, and advising the government on loans and credit facilities. The banks' minimum capital requirements vis-à-vis their risk-weighted assets were increased to 8% in most countries in the region, along the lines of the recommendations of the Basle Committee on Banking Supervision in 1995. Capital was defined to consist of two components, primary capital, which includes paid-up capital and reserves, and other capital, which includes provisions for general banking risks and subordinated long-term loans of at least five-year maturity (Murinde and Yaseen, 2004). Looking at the structure of banking sector in the non-oil countries, we can see that the state owns around 67% of banking assets in Egypt, while it owns only 29% in Morocco. However, there is no state ownership of banks in either Jordan or Lebanon. The same findings are true for both banking system loans and deposits. It is, however, worth mentioning that the number of government-owned banks in Egypt is still highest among other MENA countries.1 As for the oil countries (GCC), we can observe that they have a fairly large number of banks with an extensive network of branches. Banks in the GCC countries are financially strong and well-capitalized (Jbili et al., 1996). Most of banks in GCC are family-owned, with modest state ownership participation, although a large number of specialized banks are fully state-owned. The GCC has already set guidelines in an effort to put in place minimum requirements for banks desiring to establish branches in other GCC countries. These requirements are intended to reduce incidents of crashes and sectoral failures. Guidelines and standards have been set with respect to licensing, capital, and capital reserve, monitoring and inspection of licensed foreign banks, bank closures, minimum capital retention requirements, and a minimum age for a bank (ten years), among other requirements (Jabsheh, 2002).

1

According to World Bank database for bank regulation and supervision, by the end of 2005.

4

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

Looking at the other component of the financial sector (security markets), we can see that security markets in the MENA region attracted the attention of policymakers within the framework of developing and reforming the financial markets. Most countries in the region started reforming their security markets in the 1990s, and the reform agenda included plans to revitalize stock markets in some countries and to establish stock markets in others. Many of these countries issued new capital laws, aimed at encouraging private investment, increasing investors' protection, and enhancing the banks' role in stimulating capital markets through the establishment of mutual funds. Despite these reforms, security markets in the MENA region are still underdeveloped, with a limited number of listed companies, low free-float of shares, and thin trading. The following table presents some key financial market indicators over the 2000–06 period and compares market performances among the major MENA markets in 2000 and 2006. As seen from Table 1, panel A, the non-oil MENA countries comprise Egypt, Jordan, Tunisia, Morocco, and Lebanon. In terms of the number of listed companies, Egypt ranked first in 2006 despite the decrease in the number of such companies to 595, while Jordan ranked first in terms of the rate of increases (listed companies reached 227 in 2006 compared with only 162 companies in 2000). The number of listed companies in non-oil MENA countries totalled 949 in 2006 compared with 1348 companies in 2000, owing to the aggressive delisting procedures set by the Egyptian stock exchange to retain only companies with good transparency, disclosure, and corporate governance. In terms of market capitalization, the non-oil MENA countries witnessed an increasing market capitalization to a record of $185 billion in 2006, compared with $48.6 billion in 2000. Egypt leads the non-oil countries in terms of market capitalization ($93.4 billion) while Morocco follows with $49.4 billion. Non-oil countries recorded a total traded value of $83.5 billion in 2006, compared with only $14 billion in 2000, Egypt also leads the non-oil countries in terms of traded value ($50.2 billion) in 2006 followed by Jordan ($21.6 billion). In terms of turnover ratio, Jordan leads the non-oil countries in 2006 with ratio of 72.7% followed by Egypt's 53.7%, while the lowest turnover ratio was recorded in Tunisia, 14.3%.

Table 1 Security Markets in the MENA region. Source: MENA Countries Exchanges websites and Arab Monetary Fund, Quarterly Report, Q4 2006. Country

Turnover ratio a (%)

Number of listed companies

Market capitalization (US$ billion)

Value traded (US$ billion)

2000

2006

2000

2006

2000

Panel (A): Non-oil countries Egypt 1076 Jordan 162 Tunisia 44 Morocco 53 Lebanon 13 Total non-oil 1348

595 227 48 63 16 949

28.5 4.9 2.8 10.9 1.5 48.6

93.4 29.7 4.2 49.4 8.3 185

11.7 0.4 0.6 1.2 0.1 14

50.2 21.6 0.6 9.1 2.0 83.5

41.1 8.2 21.4 11.0 6.7 28.8

53.7 72.7 14.3 18.4 24.1 45.1

86 36 180 50 121 106 579 1528

67.9 8.2 19.8 6.6 3.5 11 117 165.6

326.9 60.9 106.0 21.1 13 167.6 695.5 880.5

17.4 0.3 4.4 0.2 0.5 0.1 22.9 36.9

1402.9 20.6 59.6 1.4 2.2 113.9 1600.6 1684.1

25.6 3.7 22.2 3.0 14.3 0.9 19.6 22.3

429.2 33.8 56.2 6.6 16.9 68.0 230.1 191.3

Panel (B): Gulf countries (GCC) Saudi Arabia 76 Qatar 22 Kuwait 86 Bahrain 41 Oman 113 UAE 27 Total GCC 365 Overall total 1713 a

Turnover ratio is calculated by authors.

2006

2000

2006

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

5

Panel B presents the development of the financial market in the GCC, where Saudi Arabia leads the GCC in terms of market capitalization ($326.9 billion) in 2006 followed by the United Arab Emirates (UAE). Total GCC market capitalization in 2006 was recorded at $695.5 billion compared with $117 billion in 2000. In terms of value traded, Saudi Arabia also leads the GCC ($1402.9 billion) in 2006, compared with only $ 17.4 billion in 2000, followed by UAE which recorded a total value traded of $113.9 billion in 2006, compared with only $0.1 billion in 2000. Saudi Arabia also leads the GCC in terms of turnover ratio, which recorded 429.2% in 2006, compared with 25.6% in 2000, and was followed by the UAE (68% in 2006), compared with only 0.9% in 2002. 3. Related literature Before embarking on our proposed research, we review the existing literature to identify how our paper could contribute to this already rich body of knowledge. The researches on bank performance were initially devoted to the analysis of bank margins. The pioneering paper of Ho and Saunders (1981) has been the theoretical framework for all empirical studies on the determinants of bank margins. The dealership model of Ho and Saunders indicates that the optimum bank interest margin depends on the bank's risk aversion, the size of bank transactions, the variance of the interest rate on deposits and loans, and the degree of market competition (see Hawtrey and Liang, 2008 and Kasma, 2010 for a detailed review of the results in both developed and developing countries). This model has been extended by Allen (1988) who introduced different types of bank products, Angbazo (1996) who augmented the model with credit risk defaults, and Maudos and De Guevara (2004) who included operating costs. An alternative approach adopted by our paper has focused on performance analysis using both net interest margins and return on bank assets and equity with a more eclectic one-step estimation procedure based on a behavioral model of the banking firm. Bank performance is usually expressed in this approach as a function of internal and external determinants. The internal variables are commonly bank specific determinants and the external variables are related to the economic, financial and institutional environment. The empirical papers on bank performance examine either cross-country or individual country banking systems. In most studies, variables such as bank size, credit risk, and equity are used as internal determinants of bank performance. Size is included to assess the existence of economies or diseconomies of scale in the banking sector. The empirical results provide conflicting evidence. Smirlock (1985), Short (1979), Bikker and Hu (2002), and Ben Naceur and Goaied (2008) find a positive and significant relationship between size and bank performance. On the other hand, Kosmidou et al. (2005) find that small UK banks display higher profitability to larger ones over the period in 1998. Kasman (2010) find that a size has a negative and statistically significant impact on the net interest margin on a panel of 431 banks in 39 countries. The relationship between equity and profitability is also controversial. The first to examine closely the capital–earning relationship is Berger (1995). The traditional view suggests a higher capital-asset ratio (CAR) is linked with a lower Return on Equity (ROE) because a higher CAR decreases the risk on equity and the tax subsidy provided by interest deductibility. More recent view based on relaxation of the symmetric information assumption claims that an increase in CAR raises ROE by reducing the expected costs for financial distress. Berger (1995) finds that an unexpected rise in capital tended to increase earning in a sample of US banks in the 1980s. He related this result to two hypotheses: first, increasing capital lower interest rate paid on unsecured debt and second, bank uses additional capital to signal that future projects are better. In more recent studies, Angbazo (1996), Demirguc-Kunt and Huinzingua (1999), Saunders and Schumacher (2000), Drakos (2003), Maudos and De Guevara (2004), Pasiouras and Kosmidou (2007), and Ben Naceur and Goaied (2008) find a positive relationship between bank performance and capitalization. In the literature on bank margins and profitability, the bank loans over total assets ratio is mainly used as a measure of bank liquidity or as a proxy for credit risk when data do not permit the calculation of the loan loss provision (Maudos and De Guevara, 2004). Miller and Noulas (1997) suggest

6

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

a negative relationship between credit risk and profitability because a higher loan to asset ratio increases the exposure of banks to bad loans and hence lowers profit margins. On the other hand, standard asset pricing arguments imply a positive relationship between risk and earnings. Empirical studies find that a higher loan ratio is associated with higher interest margins, which suggest that riskaverse shareholders seek larger earnings to compensate higher credit risk (Demirguc-Kunt and Huizingua, 1999, Chirwa, 2003, Maudos and Guevara, 2004, Ben Naceur and Goaied, 2008, and Flamini et al., 2009). However, Demirguc-Kunt and Huizingua (1999) find that the sign on loans to total assets ratio is negative in the before-tax profit over total assets equation, but when it is interacted with GDP becomes positive, indicating that at higher income level banks' lending activities tend to be more profitable. The impact of macroeconomic factors on bank performance has also been discussed in the literature. Revel (1979) was the first to suggest that the effect of inflation on bank profitability depends on whether operating expenses increase at a higher rate than inflation. Perry (1992) adds that the impact of inflation on bank profitability depends on whether inflation is fully anticipated. This implies that if inflation is totally anticipated then revenues increase faster than costs, improving in this way profitability. Most of the studies on the impact of inflation on profitability find a positive and significant relationship (Claessens et al., 2001, Bourke, 1989, Molyneux and Thornton, 1992, Athanasoglou et al., 2006, and Pasiouoras and Kosmidou, 2007). However, Afanasieff et al. (2002) and Ben Naceur and Kandil (2009) find that the inflation rate negatively affects interest margins. Afanasieff et al. suggest that inflation may be capturing the effect of seignorage collection on interest margins. Ben Naceur and Kandil explain the negative coefficient by the fact that a higher inflation rate increases uncertainty and reduces demand for credit. One could also argue that this negative relationship may be linked to slower adjustment of revenues compared with costs for inflation (Wendell and Valderrama, 2006 and Abreu and Mendes, 2003). GDP growth is also considered as a macro determinant of bank performance and allows for controlling business cycle fluctuations (Bernanke and Gertler, 1989; Kiyotaki and Moore, 1997). According to Bernake and Gertler (1989), during recessions the quality of loans declines and therefore companies borrow at higher margins, therefore a negative relationship between spread and economic growth is to be expected. Claeys and Vennet (2008) find that prevailing business cycle conditions affect net interest margins. In the Western European countries, higher economic growth is associated with higher margins, whereas in the Central Eastern European countries no link is found. The positive relationship between growth and net interest margin is also found in Schwaiger and Liebig (2008), Claessens et al. (2001) and Flamini et al. (2009). Nevertheless, a negative relationship is found in Demirguc-Kunt et al. (2004) while Dietrich et al. (2010) confirm the contercyclicality of interest margins. Using profitability indicators (returns on assets and equity), Goddard et al. (2004), DemirgucKunt and Huizinga (1998), Bikker and Hu (2002), and Flamini et al. (2009) find a positive relationship with real GDP growth. Turning to market concentration and its impact on bank profitability, it should be noted that two opposing hypotheses have been proposed: the structure-conduct-performance (SCP) hypothesis and the efficient-structure (ES) hypothesis. The SCP hypothesis states that increased market power yields monopoly powers (Short, 1979; Molyneux et al., 1996). The ES hypothesis asserts that market concentration is not the case of a bank's superior profitability and attributes the higher profit to superior efficiency that enables efficient banks to gain market share and earn higher profits (Demsetz, 1973, and Peltzman, 1977). A number of studies confirm the SCP hypothesis (Rose and Fraser, 1976, Heggestad and Mingo, 1974, Rhoades, 1977, Samad, 2005, and Chirwa, 2003). Other researches provide support to the ES hypothesis in the banking sector (Gillini et al., 1984, Smirlock, 1985, and Evanoff and Fortier, 1988) and some find no evidence to support the SCP hypothesis (Berger, 1995, Athanasoglou et al., 2008, and Ben Naceur and Goaied, 2008). The last group of profitability determinants deals with financial structure and institutional variables. Demirguc-Kunt and Huizingua (1999) find a negative relationship between the size of the banking sector and profitability measures that reflects the higher level of competition in developed banking sector. Demirguc-Kunt and Huizingua (2001) also present evidence on the impact of financial development and structure on bank performance for a large sample of countries over the 1990–97 period. The paper finds that financial development has a significant impact on bank

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

7

profitability. A developed banking system reduces profitability through higher competitiveness whereas stock market development improves bank performance especially in a lower level of financial development. The complementarity between stock market development and bank performance is also found in Ben Naceur and Goaied (2008), and Kosmidou et al. (2005). Regarding legal and institutional differences, Demirguc-Kunt and Huizingua (1999) find that better contract enforcement, an efficient legal system, and lack of corruption are associated with low profitability in a sample of 80 developed and developing countries. In the same vein, Demirguk-Kunt et al. (2004) examine the impact of bank regulation, market structure, and national institutions on bank interest margins and overhead cost using 1400 banks across 72 countries. The paper finds that bank regulation become insignificant when controlling for national indicators of economic freedom or property rights protection. Institutional development explains cross-bank differences in net interest margins. In a study on 92 countries over the period 1994–2008, Dietrich et al. (2010) find that country-level governance variables are important determinants of the internet margins with significant differences between developed and developing countries. Finally, Leaven and Majnoni (2005) investigate the effect of judicial efficiency on bank's lending spreads for a large cross-section of countries. The paper found that improvement in judicial efficiency and judicial enforcements of debt contracts are crucial for lowering the cost of financial intermediation. 4. Data and empirical model 4.1. Data We use a sample of 173 banks from ten MENA countries over the 1988–2005 period. All bank balance sheet data and income statements are obtained from the BankScope database provided by Fitch/IBCA/ Bureau Van Dijk. Since we focus on bank intermediation we use unconsolidated statements when available and consolidated statements when the unconsolidated ones are not provided, making sure that each bank is included only once in the data set. Besides, because our sample includes only commercial banks, there is homogeneity in the comparison over country groups. All bank-specific variables are calculated using the standardized global accounting format available in the Bankscope. Data on inflation, economic growth and GDP per capita are taken from the world development indicators. Data on financial development, structure, and density are from Beck and others (2007) and IFS (IMF International Financial Statistics). The International country risk guide (ICRG) database provides information on the quality of environment, such law and order and corruption indexes. As seen from Table 2, banks from non-oil countries dominate our sample banks, with Lebanon and Egypt representing 50% of the entire sample. Also, we can notice that the UAR, followed by Bahrain, dominates the banking sector in the GCC (above 50% of our sample banks). We were able to gather more data on these banks in recent years compared with earlier years as Bankscope included a limited number of banks in the 1980s, but the number kept increasing in the 1990s, reflecting developments in the banking sector in the MENA countries. 4.2. Methodology and empirical models The empirical work on determinants of bank's profitability can potentially suffer from three sources of inconsistency: highly persistent profit, omitted variables, and endogeneity bias (Poghosyan and Hesse, 2009). We adopt the dynamic panel techniques in our empirical analysis to correct for these potential problems. The linear dynamic panel data equation is specified as follows:

Perfit = β1 Perfit−1 + β2 Bik;t + β3 Ri;t + β4 Mi;t + β5 Fi;t + β6 Ii;t + βCi;t + ηi + εit

ð1Þ

Where Perfik,t is the performance of bank k in country i during the period t and is measured by three alternative measures (cost of intermediation, operating performance, and bank profitability), each

8

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20 Table 2 Distribution of the sample of commercial banks in MENA region. Country Non-oil countries Egypt Lebanon Jordan Morocco Tunisia Gulf countries Bahrain Kuwait Oman Saudi Arabia United Arab Emirates Total By year 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Total

Number

Percentage

30 60 10 10 14

16.85 33.71 5.62 5.62 7.87

12 6 9 11 16 178

6.74 3.37 5.06 6.18 8.99 100

7 7 9 59 95 113 125 132 146 149 151 153 151 141 139 138 129 126 1970

0.36 0.36 0.46 2.99 4.82 5.74 6.35 6.70 7.41 7.56 7.66 7.77 7.66 7.16 7.06 7.01 6.55 6.40 100

This table describes the sample used in our paper to investigate the determinants of the MENA banks' performance.

measure has several proxies as we indicate shortly below. Ci,t is a measure of bank concentration in country i during the period t; Bik,t is a vector of bank-specific characteristics of bank k in country i during the period t; Ri,t is a vector of regulatory impediments on banks in country i during the period t; Mi,t is a vector of macroeconomic variables in country i during the period t; Fi,t is a vector of financial development control variables in country i during the period t; and Ii,t is a vector of institutional development indicators in country i during the period t. Since our T, which refers to the number of years, is large enough (T = 16), it is more appropriate to use the system GMM estimator of Arellano and Bover (1995) and Blundell and Bond (1998). The basic ideas behind this estimator are: 1) the unobserved fixed effects ηi are removed by taking first difference in the equation, 2) the right hand side variables are instrumented using lagged values of the regressors, and the equation in first differences and in levels are jointly estimated, and 3) the validity of the instruments is tested using a Hansen-test of over-identifying restrictions and a test of the absence of serial correlation of the residuals. To estimate the regressions, we need, first, to give indication on how to measure our variables of interest: (1) bank performance indicators, (2) bank concentration, (3) bank-specific characteristics, (4) regulatory

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

9

Table 3 Summary of the variables. Variable

Definition

Panel A: Bank performance and specific variables NIM (Interest received − interest paid) / total earning assets SPREAD (Interest received / total earning assets) − (interest paid / liabilities) ROA Net income / total assets ROE Net income / equity COSTEFF Total operating costs / total earning assets + deposits LNSIZE Logarithm of total real assets CREDIT_RISK Net loans / total assets EQUITY Equity / total assets RESERVE_COST Non interest earning assets / total assets Panel B: Economic GROWTH GDPCAP INF CONC MARKET_CAP CREDIT_PRIVATE DENS LAW COR DEPINS

and institutional control variables Real GDP per capita growth Logarithm of GDP per capita Inflation rate Assets of three largest banks as a share of assets of all commercial banks Stock market capitalization / GDP Private credit by deposit money banks / GDP Total deposits of the banking sector divided by area (Km2) Law and order: A score from de 0 to 6. Low scores indicate that the law is ignored and high scores indicate a better legal enforcement. Corruption: A score from 0 to 6. Low scores indicate that the corruption is high. Coverage to deposit per capita ratio

Source Bankscope Authors' calculations Bankscope Bankscope Authors' calculations Authors' calculation Bankscope Bankscope Authors' calculation

WDI WDI WDI Beck et al. (2007) Beck et al. (2007) Beck et al. (2007) Authors' calculation ICRG (international country risk guide) ICRG Deposit insurance database

This table describes the variables used in our regression analysis to investigate the determinants of the MENA banks' performance.

policies, (5) variables to control for cross-country differences in the macroeconomic environment and (6) financial structure and development indicators, and (7) indicators of institutional development. These variables are summarized in Table 3 and are grouped under two main panels: panel A which presents bank performance and specific variables; while panel B presents economic and institutional control variables. This is what we discuss in details next.

4.2.1. Bank performance indicators We measure this performance by three alternative measures: cost of intermediation, operating performance, and profitability. Cost of intermediation: we use two proxies: net interest margin (NIM) which equals interest income minus interest expense divided by interest-bearing assets. The net interest margins measure the gap between what the bank pays the providers of funds and what the bank receives from firms and other users of bank credit. Operating performance: to measure bank operating performance. We follow Kwan (2003) and use total operating costs divided by the sum of total earning assets and total deposits2 (COSTEFF). We use operating performance in the cost of intermediation and profitability regressions as an independent variable to control for the efficiency in expenses management. Bank profitability: this is measured by the return on assets (ROA) and is calculated as the net income divided by average total assets.

2

Justified by the intermediation approach in measuring banking outputs.

10

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

4.2.2. Bank-specific characteristics We use several proxies for bank-specific characteristics as follows: Bank size: this variable is set to be equal to the logarithm of total bank assets in millions of US dollars. Size might be an important determinant of bank performance if there are increasing returns to scale in banking. However size could have a negative impact when banks become extremely large owing to bureaucratic and other reasons. Thus, we expect a non-linear relationship between size and bank performance, and to capture this relationship we use two variables: banks' real assets (SIZE) and their square (SIZE²). Bank equity: this refers to the book value of equity divided by total assets (EQUITY). Some theories (Berger, 1995, among others) suggest that well-capitalized banks are subject to less expected bankruptcy costs and hence lower cost of capital. According to this view, higher bank equity ratios may influence bank performance positively when loan rates do not vary much with bank equity. Bank risk: this is proxied by the ratio of net loans to total loans (CREDIT_RISK). We expect that a high CREDIT_RISK ratio will be associated with higher interest margins owing to risk and cost considerations. Higher CREDIT_RISK ratio should improve bank incomes since loans are the most risky and, hence, the highest-yielding type of assets. Another theory suggests that increased exposure to risk decreases profitability.

4.2.3. Macroeconomic indicators We use two proxies for macro-economic environment: inflation (INF) and GDP per capita growth (GROWTH). Previous studies have reported a positive association between inflation and bank profitability. High inflation rates are generally associated with high loan interest rates, and therefore, high incomes. However, if inflation is not anticipated and banks are sluggish in adjusting their interest rates, there is a possibility that bank costs may increase faster than bank revenues and hence adversely affect bank profitability. The GDP per capital growth is expected to have a positive impact on bank's performance according to the well-documented literature on the association between economic growth and financial sector performance.

4.2.4. Financial development indicators We also examine the impact of the level of financial development on the performance of the banking sector. We use two proxies for the level of financial development; one represents market-based indicators and the other refers to bank-based indicators. As for the first proxy, we use stock market capitalization divided by GDP (MARKET_CAP) as a measure of the size of the equity market. As for the bank-based indicators, we use the size of the ratio of the credit to private sector as a percentage of the GDP (CREDIT_PRIVATE) to measure the importance of bank financing in the economy. MARKET_CAP and CREDIT_ PRIVATE may also indicate the complementarities or substitutability between bank and equity market financing.

4.2.5. Bank concentration and density The literature contains two different positions regarding the impact of bank concentration on pricing decision and bank performance. The structure-performance hypothesis claims that a more concentrated banking sector will behave oligopolistically, while the efficient-structure hypothesis argues that concentration will be conduce to better efficiency as more efficient banks buy less efficient ones. Bank concentration (CONC) equals the fraction of bank assets held by the three largest commercial banks in the country. Bank concentration is computed using bank-level data from the BankScope database. We also compute another variable for the structure of the banking sector; that is, the density of demand (DENS) which equal the total deposits of the banking sector (obtained from the IFS database) divided by area (Km2).

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

11

4.2.6. Regulatory policies To the extent that reserve holdings are not remunerated or remunerated at less-than-market rates, these regulations impose a burden on banks. Thus, we will test whether reserve requirements impact negatively bank net interest margins and performance. Reserve requirement is proxied by the ratio of noninterest earning assets divided by total assets (COST_RESERVES). Additionally, we use coverage-to-deposit per capita ratio (DEPINS) as another variable to control for the protection provided by authorities for depositors. A better coverage will reduce the monitoring of bankform depositors, which will contribute to a decrease of bank performance.

4.2.7. Institutional constraints to competition Besides analyzing specific regulatory impediments on competition and the effect of bank concentration on interest margins, we also consider three indicators as proxies for the overall institutional environment. In particular, we investigate whether bank regulation and concentration have an incidence on bankinterest margins beyond the overall institutional environment. Empirical results suggest that better institutions boost competition throughout the economy. These studies predict that better institutional environment will have a negative impact on net interest margins (Engerman and Sokoloff, 1997). However, Bianco, Jappelli, and Pagano (2001) argue that the effect of overall institutional quality on net interest margins is theoretically not clear. As a result, the impact of better institutions on net interest margins could be ambiguous. We empirically test the incidence of overall institutional development on net interest margins and other performance measures. We use the real per capita GDP (GDPCAP) expressed in thousands of US dollars as an indicator of institutional development, since it is not easy to assess the important features of well-functioning institutions. To further control for the quality of institutions, we also include two additional variables from the ICRG data base in our regressions. The first one is the law and order (LAW) index that ranges from 0 to 6, where 0 indicates that the law is ignored and high scores indicate better legal enforcement. The second variable is the corruption (COR) index, which ranges from 0 to 6 where low score indicates that the corruption is high and vice versa.

Table 4 Summary statistics for the entire sample by variable. Variable

N

Mean

Std. Dev.

Minimum

Maximum

NIM in % SPREAD in % ROA in % ROE in % COSTEFF in % SIZE in million US $ SIZE² in million US $ CREDIT_RISK in% EQUITY in % RESERVE_COST in % GROWTH in % LNGDPCAP in US $ INF in % CONC in % MARKET_CAP in % CREDIT_PRIVATE in % DENS in US $ LAW COR DEPINS in %

1971 1971 1971 1971 1971 1971 1966 1971 1971 1971 1871 1871 1889 1875 1579 1573 1900 1777 1777 1862

3.29 3.15 1.25 12.96 1.33 1.23 32.95 43.48 11.05 10.45 1.92 8.37 4.49 55 35 361 1.43 4.25 2.31 6.22

1.76 1.95 1.99 25.09 0.91 5.60 89.04 18.79 8.33 9.62 2.84 0.94 10.50 17 35 385 1.85 0.79 0.91 21.52

−6.48 −6.01 −29.67 −312.12 −0.13 −3.79 0.00 −30.39 −42.81 0.12 −8.13 6.96 −1.35 27 4 20 0.00 1.50 1.00 0.00

16.06 25.43 30.18 547.38 13.75 25.22 635 91.67 97.41 77.03 34.62 10.08 80.74 90 220 1610 6.85 6.00 4.00 100

All country-level variables are averaged for the period 1989–2005, except Bank Concentration (Deposits) for which we use data from 1991 and institutional variables (LAW, CORRUPTION, and DEPINS) for which we use data until 2004. A detailed description of the definition and the sources of the variables are given in Table 3.

12

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

Table 5 Summary statistics for each variable by country. Variables

NIM in % SPREAD in % ROA in % ROE in % COSTEFF in % SIZE in million US $ CREDIT_RISK in % EQUITY in % RESERVE_COST in % GROWTH in % LNGDPCAP in US $ INF in % CONC in % MARKET_CAP in % CREDIT_PRIVATE in % DENS in US $ LAW COR DEPINS in %

Non-oil countries

Oil countries

Egypt

Lebanon

Jordan

Morocco

Tunisia

Bahrain

Kuwait

Oman

Saudi Arabia

UAE

1.97 1.76 1.23 12.77 0.99 2.75 46.21 9.48 6.96 2.41 7.26 5.97 0.57 0.27 0.43 0.25 3.69 2.18 0.00

3.93 4.13 0.84 14.92 1.71 3.85 28.75 9.05 16.03 2.22 8.40 7.98 0.33 0.11 0.79 3.68 3.84 1.75 0.76

3.43 3.75 0.94 12.37 1.58 8.98 43.93 7.33 16.41 2.03 7.50 2.64 0.88 0.89 0.68 0.06 4.19 3.46 3.04

5.02 4.81 0.91 7.19 1.41 1.08 50.41 9.13 12.71 1.86 7.10 2.20 0.62 0.31 0.46 0.01 5.61 3.00 0.00

3.23 3.29 0.90 9.56 1.49 6.02 65.68 11.11 12.98 3.29 7.55 3.71 0.61 0.12 0.53 0.06 4.59 2.79 0.00

2.29 1.70 1.69 9.31 1.12 11.91 37.86 15.02 4.43 2.76 9.36 0.72 0.83 0.97 0.44 2.67 5.01 3.11 4.62

1.99 1.52 1.23 9.64 0.58 24.58 34.99 11.13 3.13 0.88 9.81 1.49 0.68 0.71 0.40 0.44 4.76 2.55 100

4.34 3.74 1.52 12.98 1.54 14.37 69.85 14.17 6.13 1.98 8.97 1.28 0.69 0.18 0.31 0.01 4.70 2.92 16.24

2.95 2.65 1.57 16.57 0.87 2.71 39.55 9.97 5.88 0.77 9.12 0.77 0.56 0.46 0.23 0.00 4.85 2.00 0.00

3.72 3.07 2.46 14.81 1.18 2.30 55.25 18.66 6.19 −0.36 10.01 3.69 0.537 – – 1.37 4.00 2.00 0.00

All country-level variables are averaged for the period 1989–2005, except Bank Concentration (Deposits) for which we use data from 1991 and institutional variables (LAW, CORRUPTION, and DEPINS) for which we use data till 2004. A detailed description of the definition and the sources of the variables are given in Table 3.

Again, banks may require a lower risk contribution on their investment in countries where the law are respected and corruption is low. 5. Empirical results 5.1. Summary statistic We present summary statistics for all variables in Tables 4 and 5. Table 4 provides summary statistics for the entire sample (average for all countries), while Table 5 provides the average of each variable for each country. As we can see from Table 4, there is a clear difference among countries in which the standard deviations of most variables are quite large. This is also clear when we look at the minimum and maximum numbers. Consequently, controlling for both country- and bank-specific characteristics is of great importance in understanding the determinants of bank performance. Moving to Table 5, we can see similar trend. Average variables differ clearly among countries and this is true for non-oil as well as oil countries. In turn, controlling for country specifics leads us to more robustness results. Additionally, we can notice that correlation coefficients among variables of interest are significant in most cases (see Appendix A), so that we have to be cautious on the regression models because of the probability of any critical multicollinearity. 5.2. Results of the multivariate regression models Tables 6, 7, and 8 present regressions of net interest margin, cost efficiency, and profitability on bank-specific, macroeconomic, financial sector structure, institutional, and regulatory variables. The model seems to fit the panel reasonably well. The Wald-test indicates fine goodness of fit, the Hansentest for the validity of the over-identifying restrictions in the GMM estimation is accepted for all specifications and the second-order autocorrelation is rejected by the test for AR (2) errors. The highly

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

13

Table 6 Determinants of bank's net interest margins: GMM-in system estimation. Variables

L.NIM COSTEFF EQUITY CREDIT_RISK REALASSETS REALASSETS² RESERVE_COST

Model specifications (1)

(2)

(3)

(4)

(5)

(6)

(7)

0.522 (6.60) ⁎⁎⁎ 0.415 (3.13) ⁎⁎⁎

0.572 (5.63) ⁎⁎⁎ 0.403 (1.51) 0.020 (2.44) ⁎⁎ 0.008 (2.04) ⁎⁎

0.575 (5.86) ⁎⁎⁎ 0.372 (1.39) 0.021 (2.58) ⁎⁎ 0.006 (1.35) −0.004 (0.16) 0.001 (0.81) 0.021 (3.28) ⁎⁎⁎ −0.001 (0.11) −0.062 (4.12) ⁎⁎⁎

0.554 (8.09) ⁎⁎⁎ 0.493 (4.14) ⁎⁎⁎

0.533 (7.29) ⁎⁎⁎ 0.551 (4.68) ⁎⁎⁎

0.540 (6.98) ⁎⁎⁎ 0.568 (4.88) ⁎⁎⁎

0.528 (6.77) ⁎⁎⁎ 0.472 (3.82) ⁎⁎⁎

0.019 (2.05) ⁎⁎ 0.003 (0.98) −0.006 (0.31) 0.003 (2.41) ⁎⁎ 0.026 (3.15) ⁎⁎⁎ −0.193 (0.80) −0.537 (1.16) −1.039 (2.21) ⁎⁎ 0.031 (0.53) 0.002 (0.14) −0.000 (1.22) −0.070 (4.15) ⁎⁎⁎ 0.169 (1.70) ⁎

0.015 (1.87) ⁎ 0.003 (0.79) −0.030 (1.43) 0.004 (2.59) ⁎⁎ 0.029 (3.38) ⁎⁎⁎ 0.005 (0.43) −0.097 (4.93) ⁎⁎⁎

0.018 (2.05) ⁎⁎ −0.000 (0.06) −0.058 (1.48) 0.007 (1.17) 0.026 (2.55) ⁎⁎ 0.001 (0.06) −0.096 (3.95) ⁎⁎⁎

0.029 (3.35) ⁎⁎⁎ 0.009 (2.11) ⁎⁎

−0.544 (2.23) ⁎⁎ −0.373 (0.64)

−0.573 (1.10) −0.144 (0.16) 0.444 (0.46) 0.150 (1.51) −0.000 (0.62) 0.335 (2.63) ⁎⁎⁎

0.022 (3.41) ⁎⁎⁎ 0.009 (3.29) ⁎⁎⁎ −0.025 (1.61) 0.002 (2.00) ⁎⁎ 0.022 (3.95) ⁎⁎⁎

GROWTH INFL

−0.030 (1.59) 0.002 (2.17) ⁎⁎ 0.019 (3.04) ⁎⁎⁎ −0.000 (0.03) −0.062 (3.97) ⁎⁎⁎ −0.046 (0.37) −0.048 (0.15)

MARKET_CAP CREDIT_PRIVATE CONC DENS

0.034 (0.24) 0.038 (0.11) −0.738 (1.64) −0.064 (1.75) ⁎

GDPCAP LAW COR

0.147 (0.64) 90.98 ⁎⁎⁎

0.583 (3.29) ⁎⁎⁎ 48.11 ⁎⁎⁎

0.743 (1.79) ⁎ 34.79 ⁎⁎⁎

0.082 (1.19) −0.003 (0.68) 0.487 (0.92) 44.38 ⁎⁎⁎

1.26 16.87 1793 177

1.02 19.75 1457 153

1.04 20.65 1441 153

0.55 19.48 1292 152

DEPINS2 Constant χ² (1) – Wald AR(2) a Sargan test b Nbr. Of obs. Nbr. of banks

0.027 (0.83) 0.002 (1.59) 0.027 (2.78) ⁎⁎⁎ 0.013 (0.98) −0.039 (1.95) ⁎ −0.161 (0.54) −1.071 (1.73) ⁎

0.105 (1.73) ⁎ −0.006 (1.17) 0.824 (1.74) ⁎ 97.75 ⁎⁎

0.093 (1.04) 0.045 (2.49) ⁎⁎ −1.577 (1.81) ⁎ 133.24 ⁎⁎⁎

−0.415 (0.71) −0.037 (0.42) −0.000 (1.67) ⁎ 0.248 (1.53) 0.042 (0.51) −0.004 (0.75) 1.346 (0.78) 68.70 ⁎⁎⁎

0.54 20.06 1296 152

0.67 16.77 932 115

0.52 18.19 1292 152

0.024 (0.39)

This table presents the results from regressions conducted to determine the sources of net interest margins for MENA commercial banks. Estimations were performed using GMM dynamic model estimation in system. ⁎ t-Statistics are in parentheses and significance at the 10%. ⁎⁎ t-Statistics are in parentheses and significance at the 5%. ⁎⁎⁎ t-Statistics are in parentheses and significance at the 1%. a Test of over-identifying restrictions is asymptotically distributed as χ² under the null of instrument validity. The null hypothesis is that the instruments used are not correlated with the residuals. P-value is in parentheses. b Test for second-order autocorrelation of residuals and is distributed as N(0,1). The null hypothesis is that errors in the first difference regression exhibit no second-order serial correlation. P-value is in parentheses. A detailed description of the definition and the sources of the variables are given in Table 3.

significant coefficients of the lagged dependent variable confirm the dynamic character of the model specification. In this present study, the coefficients on the lagged dependent variables take a value of approximately 0.56 for NIM, 0.44 for cost efficiency, and 0.31 for ROA, which means that the departure from a perfectly competitive market system in the MENA banking sector is larger for net interest

14

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

Table 7 Determinants of bank's return on assets: GMM-in system estimation. Variables

L.ROA COSTEFF EQUITY CREDIT_RISK REALASSETS REALASSETS² RESERVE_COST

Model specifications (1)

(2)

(3)

(4)

(5)

(6)

(7)

0.176 (2.08) ⁎⁎ −0.860 (4.11) ⁎⁎⁎

0.373 (1.99) ⁎⁎ −0.729 (2.43) ⁎⁎

0.435 (2.58) ⁎⁎ −0.463 (1.76) ⁎

0.243 (1.88) ⁎ −0.259 (3.01) ⁎⁎⁎

0.247 (2.13) ⁎⁎ −0.259 (2.00) ⁎⁎

0.244 (1.75) ⁎ −0.274 (1.88) ⁎

0.260 (2.00) ⁎⁎ −0.231 (2.62) ⁎⁎⁎

0.163 (3.13) ⁎⁎⁎ 0.008 (3.11) ⁎⁎⁎

0.133 (2.18) ⁎⁎ 0.006 (1.56) −0.041 (1.28) 0.002 (0.95) 0.011 (1.74) ⁎ −0.004 (0.16) −0.016 (0.54) 0.490 (2.04) ⁎⁎ 0.237 (0.40)

0.084 (1.68) ⁎ 0.001 (0.30) 0.006 (0.15) −0.000 (0.10) 0.008 (1.39) 0.017 (1.72) ⁎

0.057 (4.13) ⁎⁎⁎ −0.002 (0.56) 0.043 (1.43) 0.001 (0.34) 0.008 (1.06) 0.701 (1.89) ⁎

0.004 (0.27) 0.643 (2.70) ⁎⁎⁎ −0.234 (0.48) −1.333 (1.84) ⁎

−0.053 (0.94) −0.004 (0.65) −0.409 (0.27) 259.26 ⁎⁎⁎ 1.65 ⁎

0.052 (3.74) ⁎⁎⁎ −0.003 (0.79) −0.016 (0.40) −0.010 (1.11) 0.005 (0.74) 0.009 (0.91) −0.007 (0.52) 0.876 (1.14) −0.486 (0.52) −0.852 (1.02) 0.022 (0.22) −0.000 (0.39) 0.067 (0.59) 0.008 (0.12) 0.040 (1.36) 1.680 (2.25) ⁎⁎ 8.26 ⁎⁎⁎ 1.85 ⁎

0.056 (4.11) ⁎⁎⁎ −0.001 (0.21) 0.049 (1.04) 0.001 (0.53) 0.006 (0.70) 0.006 (0.61) 0.003 (0.20) 1.261 (3.31) ⁎⁎⁎ −1.897 (3.53) ⁎⁎⁎

1.172 (1.61) 11.82 ⁎⁎⁎ 1.83 ⁎

−1.521 (2.89) ⁎⁎⁎ −2.139 (3.27) ⁎⁎⁎ −0.032 (0.63) 0.004 (0.49) −0.000 (1.63) −0.012 (1.00) −0.061 (0.75) 0.062 (1.22) −0.003 (0.82) 2.722 (4.04) ⁎⁎⁎ 13.66 ⁎⁎⁎ 1.87 ⁎

0.057 (3.19) ⁎⁎⁎ −0.002 (0.59) −0.010 (0.39) 0.003 (2.14) ⁎⁎ 0.008 (1.28) 0.015 (1.58) −0.011 (0.84) 0.177 (0.59) −0.709 (1.27)

51.93 1441 153

15.43 1292 152

16.48 1296 152

10.73 932 115

15.71 1292 152

−0.034 (1.56) 0.001 (1.07) 0.014 (2.36) ⁎⁎

GROWTH INFL MARKET_CAP CREDIT_PRIVATE CONC

−0.077 (1.11)

DENS GDPCAP LAW COR DEPINS2 Constant χ² (1)–Wald AR(2) a Sargan test b Nbr. Of obs. Nbr. of banks

−0.015 (0.03) 28.98 ⁎⁎⁎

−8.566 (0.83) 4.99 ⁎⁎⁎

1.05 32.60 1793 177

1.02 36.07 1457 153

0.006 (0.10) −0.000 (1.01)

−0.541 (0.45) −0.032 (0.27) −0.000 (0.45) −0.073 (0.76) 0.083 (1.40) 0.002 (0.22) 0.692 (0.25) 15.12 ⁎⁎⁎ 1.87 ⁎

This table presents the results from regressions conducted to determine the sources of return on assets for MENA commercial banks. Estimations were performed using GMM dynamic model estimation in system. ⁎ t-Statistics are in parentheses and significance at the 10%. ⁎⁎ t-Statistics are in parentheses and significance at the 5%. ⁎⁎⁎ t-Statistics are in parentheses and significance at the 1%. a Test of over-identifying restrictions is asymptotically distributed as χ² under the null of instrument validity. The null hypothesis is that the instruments used are not correlated with the residuals. P-value is in parentheses. b Test for second-order autocorrelation of residuals and is distributed as N(0,1). The null hypothesis is that errors in the first difference regression exhibit no second-order serial correlation. P-value is in parentheses. A detailed description of the definition and the sources of the variables are given in Table 3.

margins than for profits, and the efforts to instil competition should be focused on further freeing interest rates. Turning to the other explanatory variables, we focus in the next following sections on bank-specific effects (bank characteristics), macroeconomic and financial sector environment, and the regulatory, institutional, and concentration setting.

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

15

Table 8 Determinants of bank's cost efficiency: GMM-in system estimation. Variables

L.COSTEFF EQUITY CREDIT_RISK REALASSETS REALASSETS² RESERVE_COST

Model specifications (1)

(2)

(3)

(4)

(5)

(6)

(7)

0.440 (4.49) ⁎⁎⁎ 0.024 (3.53) ⁎⁎⁎

0.472 (2.34) ⁎⁎ 0.028 (2.36) ⁎⁎

0.443 (2.28) ⁎⁎ 0.032 (2.92) ⁎⁎⁎

0.551 (3.75) ⁎⁎⁎ 0.025 (4.39) ⁎⁎⁎

0.623 (2.91) ⁎⁎⁎ 0.027 (3.97) ⁎⁎⁎

0.647 (2.86) ⁎⁎⁎ 0.027 (3.32) ⁎⁎⁎

0.527 (3.23) ⁎⁎⁎ 0.021 (3.62) ⁎⁎⁎

0.002 (1.94) ⁎ −0.000 (0.03) −0.001 (1.57) 0.012 (3.98) ⁎⁎⁎

0.005 (2.28) ⁎⁎ −0.009 (1.16) 0.000 (0.31) 0.010 (3.50) ⁎⁎⁎ 0.003 (0.59) −0.008 (0.69) −0.073 (1.02) 0.376 (2.02) ⁎⁎

0.004 (2.59) ⁎⁎ −0.014 (1.44) 0.000 (0.46) 0.010 (3.22) ⁎⁎⁎ 0.001 (0.38) −0.016 (2.10) ⁎⁎

0.004 (3.04) ⁎⁎⁎ −0.013 (1.86) ⁎

0.004 (3.04) ⁎⁎⁎ −0.020 (2.53) ⁎⁎

0.003 (2.32) ⁎⁎ −0.049 (2.58) ⁎⁎

0.000 (0.83) 0.010 (3.21) ⁎⁎⁎ −0.025 (0.25) 0.376 (2.11) ⁎⁎

0.001 (1.03) 0.013 (3.28) ⁎⁎⁎ −0.004 (0.88) −0.011 (1.46) 0.009 (0.08) 0.442 (1.38)

0.006 (2.73) ⁎⁎⁎ −0.032 (1.48) 0.004 (0.66) 0.014 (2.91) ⁎⁎⁎ −0.010 (2.82) ⁎⁎⁎ −0.009 (1.44) 0.218 (0.94) 0.269 (1.35) −0.224 (0.36) −0.057 (0.96) 0.000 (0.59) −0.024 (0.57) 0.052 (1.45) 0.019 (1.85) ⁎

GROWTH INFL MARKET_CAP CREDIT_PRIVATE CONC DENS

−0.096 (1.22) 0.548 (2.85) ⁎⁎⁎ 0.108 (0.60) −0.016 (0.91)

GDPCAP LAW COR DEPINS2 Constant χ² (1)−Wald AR(2) a Sargan test b Nbr. Of obs. Nbr. of banks

0.209 (1.49) 32.08 ⁎⁎⁎ 0.65 20.00 1793 177

−0.017 (0.17) 30.44 ⁎⁎⁎ −0.19 38.49 1457 153

0.139 (0.65) 21.25 ⁎⁎⁎ −0.24 20.54 1441 153

−0.072 (0.40) −0.005 (0.29) −0.002 (0.38) 0.000 (0.04) −0.014 (2.29) ⁎⁎ −0.020 (0.85) 0.043 (1.49) −0.001 (0.65) −0.112 (0.61) 48.13 ⁎⁎⁎ −0.07 18.89 1292 152

−0.005 (0.32) 0.000 (0.47)

0.053 (1.41) −0.001 (0.72) −0.198 (1.14) 76.97 ⁎⁎⁎ −0.05 17.91 1296 152

−0.441 (0.82) 38.72 ⁎⁎⁎ −0.33 19.21 932 115

0.002 (2.39) ⁎⁎ 0.011 (2.77) ⁎⁎⁎ −0.004 (0.96) −0.003 (0.44) −0.071 (0.78) −0.148 (0.92) −0.330 (1.13) −0.068 (1.69) ⁎ 0.000 (0.75) −0.023 (0.83) −0.014 (0.56) −0.001 (0.61) 0.402 (0.52) 53.94 ⁎⁎⁎ −0.07 19.05 1292 152

This table presents the results from regressions conducted to determine the sources of cost efficiency for MENA commercial banks. Estimations were performed using GMM dynamic model estimation in system. ⁎ t-Statistics are in parentheses and significance at the 10%. ⁎⁎ t-Statistics are in parentheses and significance at the 5%. ⁎⁎⁎ t-Statistics are in parentheses and significance at the 1%. a Test of over-identifying restrictions is asymptotically distributed as χ² under the null of instrument validity. The null hypothesis is that the instruments used are not correlated with the residuals. P-value is in parentheses. b Test for second-order autocorrelation of residuals and is distributed as N(0,1). The null hypothesis is that errors in the first difference regression exhibit no second-order serial correlation. P-value is in parentheses. A detailed description of the definition and the sources of the variables are given in Table 3.

5.2.1. Bank characteristics The first variable is equity over total assets and the results in Tables 6–8 confirm the positive and highly significant impact of bank capitalization on net interest margin, cost efficiency, and profits. Equity is considered as an expensive financial device, so to provide a fair remuneration to stockholders, banks should provide better margins to compensate additional risks, which result in higher profits. Besides, when a bank holds a capital in excess of the regulatory minimum, two positive effects on the interest margin can be shown. Since the bank has free capital, it has the opportunity to increase its investment in risky assets in the form of loans or securities. When

16

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

market conditions enable the bank to provide additional loans with a profitable return/risk profile, this will, ceteris paribus, improve the interest margin. However, the positive impact of equity on cost efficiency is somewhat puzzling since the expected association should be negative because well-capitalized banks reflect both high- quality management and aversion to risk taking. An explanation given for this result could be attributed to the incentives provided by indebtedness to control operating costs, so if bank increase its capital the incentive disappears. Now consider credit risk measured by loans over total assets. As shown in Tables 6 and 7, bank risk enters positively and significantly in all the net interest margin and cost efficiency regressions. The positive impact of credit risk on net interest margins could be explained by two factors: banks cover their greater exposition to risk by increasing margins and the cost of loans since they need to be originated, serviced, and monitored (loans are the type of assets with the highest operational cost in a bank portfolio). As for the positive effect of credit risk on cost efficiency, it could be attributed by the increased screening and monitoring required by a higher proportion of loans in the bank's assets portfolio. On the profit side, the impact of credit risk is positive and significant only in the basic model with only the bank's characteristic variables, but the significance disappears when macroeconomic and financial variables are included. The positive sign on stock market capitalization in Table 8, equation 2, could be at the origin of this disappearance since stock market development contributes to a great extent to the improvement of transparency and hence the reduction of the screening and monitoring process of loans by banks. With respect to the cost of reserves, the results in Tables 6 and 7 suggest that the higher the reserves then the higher is the net interest margins and cost efficiency. The results also support the argument that the opportunity cost of keeping reserves, which can be considered as an implicit tax, seems to influence positively bank interest margins and cost efficiency. Thereby, commercial banks try to reflect this tax that erodes their profitability by increasing their explicit margins and passing it on to customers. Besides, the impact of the cost of reserves on profit is positive, meaning that banks make customers pay a price above the opportunity cost of keeping reserves. All estimated equations in Tables 7 and 8 show that the effect of bank size on profitability and cost efficiency is not relevant. As with the effect of size on net interest margins, Table 6 shows that the impact of size on bank margins is non-linear which mean that there is an optimum size to reach in the MENA banking sector above which diseconomies of scale should show up. The cost efficiency ratio is an important explanatory variable for interest rate margins in the MENA region. Higher industry operating costs produce higher spreads. As the theoretical model predicts, banks that support higher average operating expenses tend to generate higher margins in order to compensate their higher transformation costs and these banks pass it again on to borrowers. This behavior reflects somehow the market power of banks and the lack of competition in the lending sector.

5.2.2. The macroeconomic and financial sector environment We now turn to the effects of macroeconomic and financial structure variables. As displayed in Tables 6–8, the macroeconomic characteristics, inflation and economic growth, have differential impacts on bank margins' efficiency and profits. While real output does not appear to influence a bank's income statement, inflation shocks seem to be passed mainly through the deposit rates (see Table 6), which means that banks do not adjust their lending rates accordingly to inflation and consequently they bear the entire negative cost of inflation. In other words, banks respond to the upward adjustment in the discount rate by reducing margins, hence supporting the cost of refinancing their liquidity needs. On the other hand, inflation is associated negatively and significantly with overheads and this association contributes to the cancelling out of the negative impact of inflation on profits. The variables used as proxies for relative development of the banking sector and the stock market seem to have no impact on net interest margins in all specifications as displayed in Table 6. Next, we see in all specifications that the measure of bank development has negative signs with statistically significant coefficients in the cost efficiency regressions. This may suggest that in a well-developed banking sector banks lower their operating costs. Also, the results in Table 7 suggest that the measure of stock market development has positive and significant signs in all specifications. This suggests that banks that operate in a well-developed stock market tend to have greater profit opportunities. A possible explanation

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

17

is that the stock market contributes to the reinforcement of firm equity and thereby reduces loans problems and leads to an increase of information available on traded firms, which enable banks to better evaluate credit risk. 5.2.3. Regulatory, institutional and concentration setting First consider concentration in Tables 6–8. Bank concentration enters negatively and significantly in all the net interest margin and return on assets regressions. This outcome is consistent with Berger (1995), who, among others, supports the argument that concentration is usually negatively associated with profitability once the institutional and regulation variables are controlled for. In accordance with the theory, higher operational efficiency induces banks to pass the lower costs on to their customers in the form of lower loan rates and higher deposit rates, thereby lowering the interest margin. This explanation should be verified by the introduction of a variable measuring economic efficiency, and both efficient-structure hypotheses predict a negative relationship between interest margins and efficiency. The positive coefficient on cost efficiency in the net interest margin regressions in Table 6 is consistent with the expected association even if our variable has a negative sign since an increase in our measure of efficiency (overheads) means a deterioration of economic efficiency. As for the institutional variables, we notice in our regressions that corruption increases the cost-efficiency as well as the netinterest margins while an improvement of the law and order variable decreases the cost efficiency without affecting performance. 6. Conclusion and policy implications During the late 1980s and the 1990s several MENA countries, like many other developing countries, underwent noteworthy financial reforms, which, significantly, affected both the banking system and the domestic stock market. By reviewing the literature, it is evident that most academic studies of the impact of these reforms on the performance of financial institutions in emerging economies concentrate on large countries such as Brazil and China. However, little is known about the performance of financial institution in the MENA countries following these reforms. Using bank-level data from ten MENA countries, our study aims to assess the extent to which financial development, bank regulations, market structure, and institutional factors affect bank performance. We cover the 1989–2005 period and control for a wide array of macroeconomic, financial, and bank characteristics. The empirical results of this study find that bank-specific characteristics, in particular bank capitalization and credit risk, have a positive and significant impact on banks' net interest margin, cost efficiency, and profitability. As for the impact macroeconomic and financial development indicators exercise on bank performance, we conclude that these variables have no significant impact on the net interest margin, except for inflation. However, inflation shocks seem to be passed mainly through the deposit rates, and this means that banks bear the entire negative cost of inflation. Also, the results suggest that banks lower their operating costs in a well-developed banking sector environment (as confirmed by the negative and statically significant coefficient of the bank development variable in the cost efficient regression models). Furthermore, the stock market development variable is always positive and significant in all specifications, suggesting that banks that operate in a well-developed stock market environment tend to have greater profit opportunities. The regulatory and institutional variables seem to have an impact on bank performance as the results suggest that corruption increases the cost-efficiency and net-interest margins while an improvement of the law and order variable decreases the cost efficiency without affecting performance. The analysis suggests a clear set of policy implications for the MENA countries. It is evident that enhancing competition through easing entry of foreign banks should be accommodated since their introduction could reduce interest margins by intensifying competition. Additionally, more development in the capital markets is encouraged so as to improve transparency of banks and provide better screening and monitoring of bank activities. Also, governments should improve governance at the macroeconomic level, by, for instance, fighting corruption and better enforcing law and order as these initiatives have a positive impact on bank performance. Lastly, states are encouraged to hasten bank privatization that allows for removing ownership and control from the state to the private sector, and thus increasing competition, transparency, and performance of banks.

18

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

Appendix A

Table A1 Correlation matrix. Variables

NIM

NIM SPREAD ROA ROE COSTEFF SIZE SIZE² CREDIT_RISK EQUITY RESERVE_COST GROWTH LNGDPCAP INF CONC MARKET_CAP CREDIT_PRIVATE MARKST DENS LAW COR DEPINS

1 0.82⁎⁎⁎ 0.12⁎⁎⁎ 0.13⁎⁎⁎ 0.46⁎⁎⁎ −0.14⁎⁎⁎ −0.13⁎⁎⁎ 0.08⁎⁎⁎ 0.26⁎⁎⁎ 0.27⁎⁎⁎ −0.02 0.07⁎⁎⁎ −0.05⁎⁎ −0.21⁎⁎⁎ −0.14⁎⁎⁎ 0.21⁎⁎⁎ 0.21⁎⁎⁎ 0.07⁎⁎⁎ 0.11⁎⁎⁎ 0.01 −0.14⁎⁎⁎

SPREAD

1 0.13⁎⁎⁎ 0.13⁎⁎⁎ 0.35⁎⁎⁎ −0.15⁎⁎⁎ −0.17⁎⁎⁎ −0.01 −0.02 0.63⁎⁎⁎ 0.02 −0.03 −0.01 −0.24⁎⁎⁎ −0.20⁎⁎⁎ 0.29⁎⁎⁎ 0.31⁎⁎⁎ 0.17⁎⁎⁎ 0.03 −0.01 −0.17

ROA

1 0.40⁎⁎⁎ −0.21⁎⁎⁎ 0.01 0.02 0.06⁎⁎⁎ 0.29⁎⁎⁎ −0.08⁎⁎⁎ −0.01 0.20⁎⁎⁎ −0.05⁎⁎ 0.07⁎⁎⁎ −0.21⁎⁎⁎ −0.10⁎⁎⁎ −0.12⁎⁎⁎ −0.05⁎⁎ 0.09⁎⁎⁎ 0.02 0.01

ROE

1 −0.07⁎⁎⁎ −0.01 −0.02 −0.02 −0.05⁎⁎⁎ 0.02 0.06⁎⁎⁎ 0.05⁎⁎ 0.01 −0.06⁎⁎⁎ 0.03 −0.04⁎⁎⁎ 0.02 0.01 −0.00 0.05⁎⁎ −0.03

COSTEFF

1 −0.17⁎⁎⁎ −0.18⁎⁎⁎ −0.05⁎⁎ 0.28⁎⁎⁎ 0.21⁎⁎⁎ 0.04 −0.07⁎⁎⁎ 0.10⁎⁎⁎ −0.18⁎⁎⁎ −0.14⁎⁎⁎ 0.24⁎⁎⁎ 0.24⁎⁎⁎ 0.09⁎⁎⁎ −0.08⁎⁎⁎ 0.05⁎⁎⁎ −0.17⁎⁎⁎

SIZE

1 0.89⁎⁎⁎ 0.05⁎⁎ −0.01 −0.15⁎⁎⁎ 0.05⁎⁎ 0.29⁎⁎⁎ −0.14⁎⁎⁎ 0.47⁎⁎⁎ 0.41⁎⁎⁎ −0.17⁎⁎⁎ −0.24⁎⁎⁎ −0.09⁎⁎⁎ 0.25⁎⁎⁎ 0.26⁎⁎⁎ 0.82⁎⁎⁎

SIZE²

1 0.00 0.00 −0.21⁎⁎⁎ −0.02 0.33⁎⁎⁎ −0.09⁎⁎⁎ 0.31⁎⁎⁎ 0.34⁎⁎⁎ −0.21⁎⁎⁎ −0.27⁎⁎⁎ −0.14⁎⁎⁎ 0.22⁎⁎⁎ 0.15⁎⁎⁎ 0.94⁎⁎⁎

CREDIT RISK

1 −0.01 −0.25⁎⁎⁎ −0.00 −0.03 −0.09⁎⁎⁎ 0.30⁎⁎⁎ −0.02 −0.26⁎⁎⁎ 0.24⁎⁎⁎ −0.46⁎⁎⁎ 0.20⁎⁎⁎ 0.19⁎⁎⁎ −0.07⁎⁎⁎

EQUITY

1 −0.11⁎⁎⁎ −0.10⁎⁎⁎ 0.29⁎⁎⁎ −0.16⁎⁎⁎ 0.09⁎⁎⁎ 0.10⁎⁎⁎ −0.04 0.28⁎⁎⁎ 0.02 0.14⁎⁎⁎ −0.03 0.03

All country-level variables are averages for the period 1989–2005, except Bank Concentration (Deposits) for which we use data from 1991 and institutional variables (LAW and CORRUPTION) for which we use data till 2004. A detailed description of definitions of the variables is given in Table 3. ⁎,⁎⁎,⁎⁎⁎ Indicates significance at 10% level.

References Abreu, M., Mendes, V., 2003. Do macro-financial variables matter for european bank interest margins and profitability? Financial Management Association International. Afanasieff, T., Lhacer, P., Nakane, M., 2002. The determinants of bank interest spreads in Brazil. Working Paper. Banco Central di Brazil. Allen, L., 1988. The determinants of bank interest margins: a note. Journal of Financial and Quantitative Analysis 23, 231–235. Angbazo, L., 1996. Commercial bank net interest margins, default risk, interest rate risk and off-balance sheet banking. Journal of Banking & Finance 21, 55–87. Arellano, M., Bover, S., 1995. Another look at the instrumental variable estimation of error-components models. Journal of Econometrics 68, 29–51. Athanasoglou, P.P., Brissimis, S.N., Delis, M.D., 2008. Bank-specific, industry-specific and macroeconomic determinants of bank profitability. Journal of International Financial Markets, Institutions and Money 18, 121–136. Athanasoglou, P., Delis, M.D., Staikouras, C., 2006. Determinants of bank profitability in the South Eastern European region. Bank of Greece Working Paper No. 47. Beck, T., Levine, R., Demirguc-Kunt, A., 2007. The WB database on regulations and supervision. World Bank. Ben Naceur, S., Goaied, M., 2008. The determinants of commercial bank interest margin and profitability: evidence from Tunisia. Frontiers in Finance and Economics 5, 106–130. Ben Naceur, S., Kandil, M., 2009. The impact of capital requirements on banks' cost of intermediation and performance: the case of Egypt. Journal of Economics and Business 61, 70–89. Berger, A., 1995. The profit-structure relationship in banking: test of market-power and efficient-structure hypotheses. Journal of Money, Credit, and Banking 27, 404–431. Bernanke, B.S., Gertler, M., 1989. Agency costs, net worth, and business fluctuations. The American Economic Review 79, 14–31. Bianco, M., Jappelli, T., Pagano, M., 2001. Courts and banks: effects of judicial enforcement on credit markets. CSEF Working Papers 58. Centre for Studies in Economics and Finance (CSEF), University of Salerno. Bikker, J., Hu, H., 2002. Cyclical patterns in profits, provisioning and lending of banks and procyclicality of the new basel capital requirements. BNL Quarterly Review 221, 143–175. Blundell, R., Bond, S., 1998. Initial conditions and moment restrictions in dynamic panel data models. Journal of Econometrics 87, 115–144. Bourke, P., 1989. Concentration and other determinants of bank profitability in Europe, North America and Australia. Journal of Banking & Finance 13, 65–79.

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

RESERVECOST

1 0.07⁎⁎⁎ −0.21⁎⁎⁎ −0.01 −0.19⁎⁎⁎ −0.22⁎⁎⁎ 0.36⁎⁎⁎ 0.21⁎⁎⁎ 0.34⁎⁎⁎ −0.12⁎⁎⁎ −0.06⁎⁎⁎ −0.19⁎⁎⁎

GROWTH

1 −0.23⁎⁎⁎ 0.04 0.02 0.07⁎⁎⁎ −0.03 0.01 −0.08⁎⁎⁎ 0.11⁎⁎⁎ −0.06⁎⁎⁎

LNGDP CAP

1 −0.11⁎⁎⁎ −0.03 0.31⁎⁎⁎ −0.07⁎⁎⁎ 0.26⁎⁎⁎ 0.15⁎⁎⁎ −0.17⁎⁎⁎ 0.31⁎⁎⁎

INF

1 −0.23⁎⁎⁎ −0.20⁎⁎⁎ −0.16⁎⁎⁎ −0.08⁎⁎⁎ −0.52⁎⁎⁎ 0.01 −0.07⁎⁎⁎

CONC

1 0.63⁎⁎⁎ 0.45⁎⁎⁎ −0.51⁎⁎⁎ 0.32⁎⁎⁎ 0.50⁎⁎⁎ 0.22⁎⁎⁎

MARKET_ CAP

CREDIT_ PRIVATE

1 −0.14⁎⁎⁎ −0.17⁎⁎⁎ 0.31⁎⁎⁎ 0.33⁎⁎⁎ 0.32⁎⁎⁎

1 0.64⁎⁎⁎ −0.23⁎⁎⁎ 0.49⁎⁎⁎ −0.17⁎⁎⁎

19

DENS

LAW

COR

DEPINS

1 −0.15⁎⁎⁎ −0.54⁎⁎⁎ −0.13⁎⁎⁎

1 0.25⁎⁎⁎ 0.18⁎⁎⁎

1 0.09⁎⁎⁎

1

Chirwa, E.W., 2003. Determinants of commercial banks' profitability in Malawi: A cointegration approach. Applied Financial Economics 13, 565–577. Claessens, S., Demirgüc-Kunt, A., Huizinga, H., 2001. How does foreign entry affect domestic banking markets? Journal of Banking & Finance 25, 891–911. Claeys, S., Vander Vennet, R., 2008. Determinants of bank interest margins in Central and Eastern Europe: a comparison with the west. Economic Systems 32 (No. 2). Creane, S., Goya, R., Mobarak, M., Sab, R., 2003. Financial development and economic growth in the Middle East and North Africa Finance and Development 4 (No. 1). A quarterly magazine of the IMF. Demerguç-Kunt, A., Huizinga, H., 1999. Determinants of commercial bank interest margins and profitability: some international evidence. World Bank Economic Review 13, 379–408. Demerguç-Kunt, A., Huizinga, H., 2001. Financial Structure and Bank Profitability in Financial Structure and Economic Growth: a Cross-country Comparison of Banks, Markets, and Development. MIT Press, Cambridge, MA. Demirguc-Kunt, A., Huizinga, H., 1998. Determinants of commercial bank interest margins and profitability: some international evidence. Policy Research Working Paper Series 1900. The World Bank. Demirgüç-Kunt, A., Laeven, L., Levine, R., 2004. Regulations, market structure, institutions, and the cost of financial intermediation. Journal of Money, Credit, and Banking 36, 593–622. Demsetz, H., 1973. Industry structure, market rivalry, and public policy. Journal of Law and Economics 16, 1–9. Dietrich, A., Wanzenried, G., Cole, R.A., 2010. Why are Net-Interest Margins Across Countries so Different? Available at SSRN: http:// ssrn.com/abstract=1542067. Drakos, K., 2003. Assessing the success of reform in transition banking 10 years later: an interest margins analysis. Journal of Policy Modeling 25 (3), 309–317. Engerman, S., Sokoloff, K., 1997. Factor endowments, institutions, and differential paths of growth among new world economies. In: Haber, S. (Ed.), How Latin America Fell Behind. Stamford University Press, Stamford, CA, pp. 260–304. Evanoff, D.D., Fortier, D.I., 1988. Reevaluation of the structure — conduct performance paradigm in banking. Journal of Financial Services Research 313–329 I (June). Flamini, V., McDonald, C., Schumacher, L., 2009. The determinants of commercial bank profitability in sub-saharan Africa. IMF Working Paper 09/15. International Monetary Fund, Washington. Gillini, T., Smirlock, M., Marshall, W., 1984. Scale and scope economics in the multi-product banking firm. Journal of Monetary Economics 13, 393–405. Goddard, J., Molyneux, P., Wilson, J.O.S., 2004. The profitability of European banks: a cross-sectional and dynamic panel analysis. Manchester School 72 (3), 363–381.

20

S.B. Naceur, M. Omran / Emerging Markets Review 12 (2011) 1–20

Hawtrey, K., Liang, H., 2008. Bank interest rate margins in OECD countries. North American Journal of Economics and Finance 19, 249–260. Heggestad, A.A., Mingo, J.J., 1974. Prices, Nonprices, and Concentration in Selected Banking Markets. Proceedings of a Conference on Bank Structure and Competition. Federal Reserve Bank of Chicago, Chicago, pp. 69–95. Ho, T., Saunders, A., 1981. The determinants of bank interest margins: theory and empirical evidence. Journal of Financial and Quantitative Analysis 16, 581–600. Jabsheh, F.Y., 2002. The GATS Agreement And Liberalizing The Kuwaiti Banking Sector. ERF, 8th Annual Conference, Egypt. Jbili, A., Galbis, V., Bisat, A., 1996. Financial Systems and Reform in the Gulf Cooperation Council Countries. Paper presented at the Workshop on Financial Market Development, Arab Monetary Fund and the Economic Research Forum for the Arab Countries, Iran and Turkey, Abu Dhabi, UAE, 25–27 May. Kasman, A., 2010. Consolidation and Commercial bank net interest margins: evidence from the old and new European union members and candidate countries. Economic Modeling 27, 648–655. Kiyotaki, N., Moore, J., 1997. Credit cycles. Journal of Political Economy 105, 211–248. Kosmidou, K., Pasiouras, F., Tsaklanganos, A., 2005. Factors Influencing the profits and size of Greek banks operating abroad: a pooled time-series study. Applied Financial Economics 15, 731–738. Kwan, S.H., Liebig, D., 2008. Operating performance of banks among Asian economies: an international and time series comparison. Journal of Banking & Finance 27 (3), 471–487. Leaven, L., Majnoni, G., 2005. Does judicial efficiency lower the cost of credit? Journal of Banking & Finance 29, 1791–1812. Maudos, J., Fernandez de Guevara, J., 2004. Factors explaining the interest margin in the banking sectors of the European union. Journal of Banking & Finance 28, 2259–2281. Miller, S.M., Noulas, A.G., 1997. Portfolio mix and large-bank profitability in the USA. Applied Economics 29 (4), 505–512. Molyneux, P., Thorton, J., 1992. Determinants of European bank profitability: a note. Journal of Banking & Finance 16, 1173–1178. Molyneux, P., Thornton, J., Lloyd-Williams, D.M., 1996. Competition and market contestability in Japanese commercial banking. Journal of Economics and Business 48, 33–45. Murinde, V., Yaseen, H., 2004. The Impact of Basle Accord regulations on bank capital and risk behaviour: Evidence from the Middle East and North Africa (MENA) region. Third International Conference of the Centre for Regulation and Competition (CRC), on ProPoor Regulation & Competition: Issues, Policies and Practices. CapeTown, 7–9 September. Pasiouras, F., Kosmidou, K., 2007. Factors influencing the profitability of domestic and foreign commercial banks in the European union. Research in International Business and Finance 21, 222–237. Peltzman, S., 1977. The gains and losses from industrial concentration. Journal of Law and Economics 20, 229–263. Perry, P., 1992. Do banks gain or lose from inflation? Journal of Retail Banking 14, 25–40. Poghosyan, T., Hesse, H., 2009. Oil prices and bank profitability: evidence from major oil-exporting countries in the Middle East and North Africa. IMF Working Paper 09/220. International Monetary Fund, Washington. Revell, J., 1979. Inflation and financial institutions. Financial Times, London. Rhoades, S.A., 1977. Structure-performance studies in banking: a summary and evaluation. Staff Economic Study No. 92. Board of Governors of the Federal Reserve System, Washington, DC. Rose, S.A., Fraser, D.R., 1976. The relationship between stability and change in market structure: an analysis of bank prices. Journal of Industrial Economics 24, 251–266. Samad, A., 2005. Banking structure and performance: evidence from Utah. Review of Business Research V (2), 151–156. Saunders, A., Schumacher, L., 2000. The determinants of bank interest rate margins: an international study. Journal of International Money and Finance 19, 813–832. Schwaiger, M.S., Liebig, D., 2008. Determinants of bank interest margins in Central and Eastern Europe. Financial Stability Report 14, 68–87 (Österreichische Nationalbank). Short, B.K., 1979. The relation between commercial bank profit rates and banking concentration in Canada, Western Europe and Japan. Journal of Banking & Finance 3, 209–219. Smirlock, M., 1985. Evidence on the (non) relationship between concentration and profitability in banking. Journal of Money, Credit, and Banking 17, 69–83. Wendell, S., Valderrama, L., 2006. The monetary policy regime and banking spreads in Barbados. IMF Working Paper 06/211. International Monetary Fund, Washington.