On Bank Market Structure and Competition in Egypt Bank Market Structure and Competition in Egypt Mahmoud Mohieldin1 1 The author wishes to thank Ziad Bahaa El-Din, Hany Dimian, Pal

On Bank Market Structure and Competition in Egypt

Mahmoud Mohieldin1

1 The author wishes to thank Ziad Bahaa El-Din, Hany Dimian, Pal Gaspar, Rolf Luders, Sahar Nasr, MohamedOzalp and the participants of the conference on “Financial Development and Competition in Egypt” for usefulcomments and discussions. He is also grateful to Rasha Awad for research assistance. Usual disclaimers apply.

1

“Why is the industrial structure as itis? …[S]ome parts of industrial structure aredetermined by factors outside the influence ofparticipating firms… [N]umber of firms, …,are seen as determined by a series of historicaccidents which are not directly relevant tocurrent conduct and performance.”

Malcolm Sawyer (1981), p. 154.

Abstract

This paper outlines the evolution of the Egyptian banking system since 1856 in an attemptto understand the impact of development and financial policies during the different phases on thebank market structure and competition. The structure-conduct-performance model (SCP) was usedto test the relevance of the structure-conduct-performance paradigm (SCP) in the Egyptian bankingcontext using pooled cross-section time series data specifically collected from 47 commercial banks,out of a total of 59 commercial banks. The paper shows that the Egyptian banking system during the1980s and 1990s faced a mixture of both the domination of public banks and a high degree ofconcentration which discourages competition on one hand, and the heavy barriers to entry whichmakes incumbent banks incontestable on the other. However, recent financial liberalization measureseliminated interest rate controls, eased the entry of new financial intermediaries, and allowed newtypes of instruments. Regulatory impediments were relaxed and the Egyptian banking systembecome more integrated with the world market. Further reform measures are required at two levels:first, at the banking units level by improving their management, granting them adequate autonomyin decision making, and encouraging training schemes and the acquisition of essential resources forefficient banking. And second, at the banking system level, by encouraging competition andcontestability, within an adequate regulatory environment, removing excessive barriers to entry andreplacing them with an objective entry criteria and establishing a reliable exit mechanismaccompanied with an efficient deposit protection scheme.

2

I. Introduction:Despite the partial liberalisation measures of the 1970s that encouraged the establishment of foreign,private and joint-venture banks and the more comprehensive financial reform program of the 1990s,branches of the public sector commercial banks dominate the Egyptian banking system. The structureof the banking system and the geographic concentration of branches indicate a highly segmentedmarket and lack of competition. The relatively large branch networks of a few public sector banksallow them to dominate the process of savings mobilisation from the public. Other banks targetmainly big savers. Financial services are still basic with very limited innovation, though joint-venturebanks are in a better position. The domination of public banks in a highly concentrated marketresulted in a frail competition and limited innovation. It is not just lack of competition that the Egyptian banking system is suffering from, it is also lackof contestability. It is argued that contestable markets and potential firms' freedom of entry, promoteefficiency, encourage innovation and give highly favorable welfare outcomes. This can be achievedby efficient pricing and allocation of production among incumbent firms to eliminate significantexcess profits. However in practice there are different barriers to entry facing potential firms andpreventing them from joining the market even in the presence of high excess profits.

Large economies of scale and high sunk costs, in addition to other entry costs are examples of suchbarriers, but in the case of banking government regulations through permits and licenses are far moreimportant than other barriers. However, the recent financial liberalisation eliminated interest ratecontrols, eased the entry of new financial intermediaries, and allowed new types of instruments.Regulatory impediments were relaxed and the Egyptian banking system became more integrated withthe world market.

The study attempts to investigate the impact of the structure of the Egyptian banking system on bankperformance and examines the evolution of the banking market and identifies the factors that formedits structure and the extent of competition, using various indices of market concentration andperformance measures.

The chapter starts with a discussion of the evolution of the Egyptian banking system since 1856 inan attempt to understand the impact of development and financial policies during the different phaseson the bank market structure and competition. The development of the Egyptian banking system canbe divided into four main phases. The first starts in 1856 with the establishment of the first bank inEgypt and ends in 1956 with the start of the Egyptianisation of foreign banks. The analysis of thisparticular period is important for three main reasons. First to understand the distinctivecharacteristics of the banking system in Egypt that shaped its structure and role before the 1952revolution and after. Second most of the operating banks in the subsequent periods, including thecentral bank, were established during this phase which requires some discussion of their historicalbackground. Third, many of the measures taken concerning banking activities in the 1950s and 1960swere, to a great extent, reactions to policies and practices adopted during this phase and affected thecontemporary development and progress of the Egyptian banking system.

3

The second phase 1957-74 was a period of heavy government intervention in economic activity andwitnessed the Egyptianisation and then the nationalisation of banks and the official adoption offinancial repression through various measures.2 The third phase, which we consider to be a periodof partial liberalisation, starts in 1974 with the introduction of open door policy (infitah) and endsin 1991 with the start of the application of the economic reform programme supported by the WorldBank and the IMF, which marks the beginning of the fourth phase.

The chapter proceeds by providing discussion of the structure-conduct-performance paradigm in theEgyptian context. Empirical tests of the structure-conduct-performance paradigm in banking industryhave often found that the structure of the market, using market concentration as a proxy, hasinfluenced both conduct and performance. Then, we analyse the relationship between performanceof banks measured by ROA, as a dependent variable, and a proxy for market structure (MCR) andother independent variables that include both bank and overall market specific variables. Weformulate a cross-sectional profit equation similar to those used in previous studies that followedWeiss (1974) and Smirlock (1985). More recently, the same approach has been used in a study on the effects of financial liberalisation on market structure in Turkey by Denizer (1997) and in anotherone on the relationship between market structure and performance in the Jordanian banking systemby Al-Karasneh, Cadle and Ford (1997). Finally. The chapter ends by a comment on the main policyimplications.

II. The evolution of the Egyptian Banking Structure:II.1 The first phase: 1856-1956Liberal banking under foreign control:

Commercial banks were the first financial institutions to appear in the process of economicdevelopment in Egypt. Their establishment in the second half of the 19th century were closelyassociated with the needs of the extensive cultivation of cotton. Although cotton started to have animportant role in the Egyptian agricultural sector from the beginning of the 19th century3, it did notreceive considerable attention until the time of the American Civil War (1861-1865). Americanexports of cotton fell during this war, giving an immense motivation to cultivate it in Egypt. Cottonproduction doubled and its price increased four times during the 1860s. It became the prime Egyptianproduct and the main source of exports revenue for the following decades, until oil started to takeover its position in the 1970s. The opening of the Suez Canal in 1869 as an important waterwaylinking the east with the west emphasised the potential importance of Egypt in international trade.

During this period, there was a chronic shortage of capital that was hindering further growth in theEgyptian economy and its trade. Agricultural borrowers were relying on village money lenders.Foreign trade was receiving insufficient funding from a few finance houses similar in their operationto the old merchant bankers of the City of London. There was a manifest need for an organised

2 For a detailed analysis of the causes, measures and impact financial repression in Egypt, see Mohieldin (1995).3 Cotton was introduced by Mohammed Ali the Ruler of Egypt (1805-1848). For an overview of this period see Owen

(1981), pp. 64-82 and Vatikiotis (1991), pp. 49-69.

4

formal financial system. Developing such a system was hindered for several reasons. Among themwas the inability of banks to introduce financial services that do not break the Islamic ban on usury;high financial instability during this period; the modest per capita income, and hence lack of savings;and the limited experience in credit dealings. These very reasons, after more than 100 years, stillimpede the development of the Egyptian financial system. However an attempt was undertaken toestablish a government-owned commercial bank in 1830 with a capital of £300,000. This attemptfailed, simply because the government could not raise the capital required.

The dearth of finance in Egypt received the attention of foreign financiers who found in the countryan attractive outlet for their capital. Foreign capital started to flow towards the country in the formof loans to the government. In 1862 Egypt had its first foreign loan in its modern history.4 Borrowingabroad continued by successive governments in order to finance the repayment of old debts and fundits projects.

Foreign debt accumulated, and the country was saddled by a debt burden which it could not manage.Foreign control was imposed on Egypt to manage the repayment of the debt. Thus the Caisse de laDette Publique, was created in 1876 as an exchequer office for the government and was entirelycontrolled by a commission of foreign creditors. Its objectives were mainly to collect public revenue,service the public debt and engage, at the same time, in commercial transactions.

Foreign financier's response to the growing need for capital in Egypt took also the form ofestablishing local financial institutions. Thus modern foreign-owned commercial banks werefounded to fund the cultivation of crops, mainly cotton, and finance foreign trade. Some of thesebanks were branches of European banks and a few of them were registered in Egypt. The first ofthese banks was the Bank of Egypt which was established in 1856 with a head office in London, amain office in Alexandria and a branch in Cairo. There is no agreement between the sources on thesize of the capital of this Bank. However it is variously reported to be between £500,000 -£1,000,000and the paid-up capital was £250,0005. Its claimed objective was to encourage trade between Egyptand Britain. The bank was also involved extensively in the purchase of Treasury Bonds which wereissued by the government to finance its current spending.

Many foreign banks were established in the following years. Their main activity was the purchaseof Treasury Bonds and meeting the increasing credit demands of the khedive. However the debtcrisis of the 1870s, the associated establishment of the Caisse de la Dette and, then, the putting ofEgyptian finance under strict foreign control in 1879, made it difficult for several banks, especiallythe new ones, to continue. Thus many of these banks were short lived and were liquidated, theremaining ones limited their activities to small scale operations6.

Towards the end of the 19th century, it was realised that the Caisse de la Dette, would disappear at

4 . The loan was for £E 3,292,800 million at 7% and redeemable in 30 years5 See NBE, op. cit., p. 13.

6 See Shafi'i (1985), pp. 183-185 and NBE (1948), op. cit., p. 13.

5

some stage, after solving the debt problem7. On the other hand the financial development of thecountry was paralysed as there were few operating foreign banks and no Bank of Issue. Thus it wasfelt that "an institution of the nature of a State Bank would be necessary to carry on the Treasurybusiness of the Government and the work done,.., by the Caisse in respect of the Public Debt"8.

Hence the National Bank of Egypt (NBE) was established in 1898 as a commercial bank with acapital of £1 million, with the head office of the bank was in Cairo. However the bank was ownedand managed by British citizens and maintained close ties with England through what was knownas the London Committee. This committee was supposed to be of consultative nature but it had a sayon transactions exceeding £100,000 and other important decisions.

According to the Khedival decree9 the NBE was granted, exclusively, the privilege of issuingbanknotes, which were introduced for the first time in the Egyptian monetary system.10 The NBEacted also as the bank of the government, a private note-issuing bank, and financial adviser to thegovernment. But the NBE was not functioning as a bankers' bank, last resort lender or controller ofcredit supply. As a commercial bank, the NBE participated as well with the rest of foreigncommercial banks to provide short-term finance for big cotton cultivators.

The operating banks benefited from the cotton-related boom of the 1902-1907 period. The value ofexported cotton increased by 52% during these six years due to a sudden rise in the internationalmarket. This favourable shock was accompanied by political stability, which stemmed from theAnglo-French agreement of 1904. Consequently the economy observed unprecedented inflows ofcapital mainly in the form of mortgages secured by agricultural and urban land. To sum up, thisperiod started with excessive speculation in land, shares and cotton and ended with an inevitablecrash.

Many banks faced bankruptcy and closure in 1907, others were left very vulnerable to any furthershocks in the future. Thus the Bank of Egypt was forced into liquidation in 1911 when it facedfurther problems. It could not call in mortgages on land to pay short term bills borrowed fromLondon. This was a result of the bank's unsound practice of short term borrowing and investing inlong term loans secured on agricultural and urban properties11. More commercial foreign banks were 7 As a matter of fact the NBE did not take over the responsibility of the Caisse de la Dette until 1940 when the

government reached an agreement with the French and the British government regarding the abolishing of the latter.NBE (1948), op. cit., p. 69.

8 Quoted in NBE (1948), p. 16.

9 Decree of 25th June 1898.

10 It is worth noting that before 1885 there was no uniform currency. The Egyptian monetary system was based on goldand silver as a standard of value and medium of circulation, and several Turkish and European currencies were incirculation. In 1885, a reform law was passed defining the currency unit in terms of gold only, allowing the Britishsovereign to be the chief instrument of circulation and preventing dealings in any other foreign currency. For furtherdiscussion of the development of the Egyptian monetary system during this period, see Shafi'i, (1985), pp. 175-207.

11 See NBE (1948), p. 34.

6

established in the following years. Many of these banks focused in serving some foreigncommunities like the Greek banks. However two aspects were shared by foreign banks. Firstgeographic concentration, as these banks were located in the capital and main ports. Second theyeither financed large cotton cultivators or foreign trade.

Effectively the country was still deprived of a banking system, which would contribute to its broaderdevelopment. Financing crops other than cotton was neglected. Advancing loans to small cultivatorswas considered either risky or technically difficult. Funding industrial projects was not compatiblewith the policy of short term financing adopted by foreign banks. Attempts to change this passiverole of the operating banks were undertaken and took the form of government intervention andprivate initiatives. They were inspired by the independence movement and marked the maindevelopments of the banking system until 1952, and took the form of extending agricultural creditand funding industrial projects.

II.1.1 Agricultural credit:Owing to the lack of acceptable collateral and high administrative costs, foreign banks generallyabstained from lending to small cultivators who continued to borrow from private moneylenders,usually brokers and merchants acquainted with local farmers. The literature usually describesinformal borrowing as being at high exploiting interest rates and at unfavourable terms, whichnecessitates intervention. We argue that there are two reasons that justify this argument in Egypt.First moneylenders used to enjoy monopolistic positions in their respective villages, with theexception of some competition of cotton exporters, who used to advance credit to large cultivators,and the Sugar Company in the south, which financed the growers of sugar cane in Upper Egypt. Therest of the credit market was left to moneylenders. Second, many of these moneylenders wereactually working as intermediaries between the foreign banks and their clients in rural areas, whichmeans that they added a margin on the bank interest rates. Further it is reported that suchmoneylenders were charged higher interest rates than other bank clients, e.g. cotton exporters. In atypical year they were charged 5-6.5 per cent while cotton exporters were charged only 4.5 per cent.12

The Agricultural Bank of Egypt was established in 1902, by the collaboration of the government andthe NBE. The establishment of the Agricultural Bank was based on an earlier successful experiencewhen the government stepped in, during the 1890s, to provide peasants with loans at relatively lowinterest rates. The Agricultural Bank and the NBE maintained close links. Approximately one thirdof the capital of £1.25 million of the Agricultural Bank, was subscribed by the NBE. TheAgricultural Bank's objective was to assist small cultivators, defined then as those who own 5feddans of land or less. The bank progressed steadily in due course. Its capital increased three timesin 10 years and its working capital was soon fully employed.

After the crisis of 1907 and the accompanying instability in the land market, the governmentintervened to protect small cultivators and enacted in 1911 the 'Five Feddan Law' that forbade theforeclosures of the owners of five feddans of land or less. Such immunity from seizure for debts didprotect small landowners but, at the same time, it had a negative impact on their financing 12 See Issawi (1963), p. 260, footnote 1.

7

arrangements. They could not obtain credit from the Agricultural Bank as they were unable toprovide the required collateral. Small cultivators and landowners again had to rely on localmoneylenders.13 Moreover, the Agricultural Bank, as a result of the Five Feddan Law, lost 85% ofits business. The bank struggled during the subsequent years with a continuing decline of itsactivities as the total amount of loans fell from £E 7 million in 1911/12 to as low as £E 0.5 millionin 1931/32. This led eventually to the liquidation of the bank in 193614.

The predominance of small scale farming and the evident requirement for relatively cheap short termcredit more than any other form, emphasised, again, the need for a financial institution which catersfor small cultivators as well as agricultural co-operative societies. Thus the Crédit Agricole d'Egyptewas established in 1931 with a capital of £E 1 million, of which the government subscribed one halfand the other principal banks the other. The bank was exempted from the Five Feddans Law andutilised the services of tax collectors to recover loans which enabled the bank to reduce costs andensure repayment. Interest rates charged were quite modest ranging between 1% and 2.5%.15 Thebank was able to charge such low interest rates thanks to a special appropriation, of £E 9 million,provided by the government at a low rate of 1%.

The bank developed rapidly, and by 1940 it had over 100 branches and 500 barns distributed all overthe country. It extended its activities to supply fertilisers and collect grain and increased its advancesto co-operatives from 5% in 1933 to 29% in 1940. In 1948 the capital of the bank increased by £E0.5 million, half of which was subscribed by the government and the rest by rural and consumer co-operatives. The bank's name changed to Agricultural and Co-operative Credit Bank to emphasise itsco-operative function. This function developed as the bank gradually channelled its credit throughthe co-operative societies which increased from 38% in 1956 to 50% in 1958 and 84% in 1960.16

II.1.2 Industrial credit:Until 1920 several foreign banks were operating in Egypt in addition to the NBE which was legallyconsidered an Egyptian bank, despite the fact that it was founded with British capital. The conditionsof the First World War and the outbreak of the Egyptian revolution in 1919 inspired a campaign toestablish a pure Egyptian bank as a necessary element of economic independence.17 It was realisedthat the operating banks with their concentration on short term financing for cotton and foreign tradedid not cater for the growing needs for long term industrial credit. Thus, Banque Misr was founded 13 See Hansen (1991), p. 52.

14 NBE (1948), op. cit., pp. 26-27.

15 See Issawi (1963), p. 260-261.

16Issawi, op. cit. p. 262.

17 The idea of establishing a pure Egyptian bank dates back to the period following the crisis of 1907 and was discussedin the Parliament in 1911 which passed a recommendation of establishing such a bank but without success. It was TalatHarb Pasha, a leading economic figure who adopted the idea and started a campaign in 1919 to establish the first pureEgyptian bank which was officially inaugurate in May 1920. (see Deeb (1976), p. 70, and for a discussion of the mainpolitical developments of this period and its economic implications see Tignor (1976) and Davis (1983).

8

in 1920 with an initial capital of £E 80 thousand raised to £E 1 million under a precise condition thatonly Egyptians could be shareholders and members of the board of directors. The capital of the bankwas subscribed by 126 shareholders; mainly large landowners and big merchants who benefited fromthe economic boom which followed the World War I.18

In ten years the volume of deposits increased from £E 200 thousand to £E 7 million in 1930 to animpressive total of £E 54 million in 1947 reflecting the confidence of the public and the progressingfunction of the bank which became second only to the NBE19. It is worth noticing that the bankreceived support from the government which took the form of annexing Post Office Savingsaccounts to it in 1927. By this move the contribution of small deposits in financing the activities ofthe bank increased remarkably till it reached a peak of approximately 75% of total deposits in 1939.

Banque Misr was influenced by the German banking system and its role in the rapid expansion ofGerman industry and trade after 1880. The concept of a gross-bank, i.e. universal bank in currentterminology, inspired the founders of Banque Misr to apply this idea in Egypt20. However we candistinguish between two main methods adopted by Banque Misr in supporting domestic industry.

− First, it was able to take direct participation in the ownership of companies. Through itsaffiliated companies of Misr Group, the bank played an impressive role in the earlyindustrialisation of the country. The number of these companies reached 27 in 1940 rangingfrom dairy products, fisheries to spinning and weaving, insurance to airlines covering theMiddle East region. The aggregate paid-up capital of the affiliated companies was £E 20million.21

− Second, it was able to provide loans and advances. In addition to participation in their capital,Banque Misr advanced credit to some of its affiliated companies. For example in 1936, sevenMisr companies received £E 1.5 million. Further, the bank borrowed from the government from1932 to 1939 an aggregate amount of £E 1.1 million mainly to finance its affiliated companies.22

However, the massive spread of Banque Misr's activities involved high risks and the bank startedto face difficulties in the late 1930s. One of Banque Misr affiliated companies was the SociétéFonciére d'Egypte which was taken over by the bank in 1927. This financial company, following anagreement with the government in 1930, took over the debts of landlords who were faced withforeclosure as a result of their inability to pay mortgages secured on their lands. Due to thecontinuing drop in cotton prices, outstanding loans were not paid to the Société Fonciére and BanqueMisr, as a holding company, was held responsible.

18 Davis (1983), op. cit., pp. 108-109 and Deeb (1976), pp. 70-71.

19ibid., p. 145.

20ibid., p. 97.

21Issawi (1963), p. 274.

22Ibid., p. 265.

9

Moreover, due to the bank's continued practice of short term borrowing and excessive long termlending, the bank suddenly found itself facing a run by numerous depositors of the Post OfficeSavings accounts and other small depositors when the hostilities of World War II started in 1939.The problem was severe as small deposit accounts comprised 75% of the Bank's total deposits of £E12.7 million. However the government, under public pressure, intervened to alleviate the bank'sdifficulties by guaranteeing its deposits and ordering the NBE to provide Banque Misr with therequired liquidity.23

The new management of the bank, which took over after the crisis of 1939, for political reasons, wasnot keen to continue the support of the bank that was competing with foreign capital within theEgyptian economy. It was concerned mainly with extracting profit from the bank and its affiliatedcompanies. The distinctive role of Banque Misr started to vanish and the bank effectively lost itsraison d'être.24

However, after the end of the Second World War the need for industrial credit to small and mediumsize enterprises was again recognised, especially when Banque Misr ended up acting like the rest ofcommercial banks. Thus in 1947 the Industrial Bank was established with 51% of its capital ownedby the government, 30% subscribed by the operating banks and 20% were in form of shares held bythe public. The government guaranteed a minimum of 3.5% of the nominal values of the shares asprofits. The aim of the bank was to widen the industrial base by establishing small enterprises oradvance credit to them. Until early 1950s the bank was not able to fulfil its objectives. It advancedcredit to relatively big projects which were able, anyhow, to obtain credit from commercial banksand was reluctant to advance credit to newly established small projects because of high riskassociated with their activities.25

During the early years after the 1952 revolution the new government was trying to assure foreigninvestment about the stability of the economy. Thus the banking system did not observe any unusualchanges regarding its structure, activities or the ownership of its institutions.26 Nevertheless the SuezCrisis of 1956 led to the subsequent radical measures of Egyptianisation by the government, underNasser, in 1957.

23 It worth noting that the NBE was reluctant at the beginning to assist Banque Misr as there was political conflict

between the management of the two banks. The government under public pressure agreed to help Banque Misr out fromthis crisis with a condition that Harb Pasha, the founder of the Bank, should resign from his position as a managingdirector. See Shafi'i (1985), op. cit., p. 315 and Davis (1983), op. cit 166-169.

24 Deeb (1976), p. 79.

25Doghaim (1989), pp. 135-137.

26 For an analysis of economic and institutional changes during this period see Mabro (1974), chapter (6).

10

II.2 The second phase 1957-1974:The adoption of financial repression:We have shown above that there were some Egyptianisation attempts before the 1952 revolution.The most important of these were the establishment of Banque Misr in 1920 and the Egyptianisationof the management and capital of the NBE during the 1940s.27 However, the measures ofEgyptianisation, established by Law 22/1957, were far more substantial than any thing that wentbefore. The law stipulated that all operating British and French banks should be sequestrated. Therest of the operating banks had to take the form of joint stock companies, within five years. The paid-up capital of operating banks should not be less than £E 500 thousand, in the form of shares ownedby Egyptians.

Small banks, which could not fulfil the capital requirements under the new law, either joined one ofthe Egyptian banks or closed down. As a result of these measures the number of banks decreasedfrom 35 in 1957 to 27 in 1958.28

The government felt that it had to assume more control over the credit market, so the NBE as aCentral Bank was granted more power by the Law No. 163 of 1957. Determining interest anddiscount rates now became the right of the NBE29. We mentioned above that setting interest rateswas left previously to the League of Egyptian Banks, which succeeded the Conférence des Banquesin 1953. The bank's supervisory authority was confirmed, as registration of banks, the opening ofnew branches and mergers were all put under its charge. The NBE was entitled as well to use reserveand liquidity ratios to control the credit market.

It was recognised that some of the functions of the League of Egyptian banks conflicted with thoseof the NBE as a Central Bank. Hence the law tried to organise the structure and role of the former.First it made its membership limited to commercial banks only. Second, it put a condition that allthe decisions of the league should be authorised by the NBE. Consequently the role of the newLeague of Commercial Banks in the operations of the banking system and monetary policy becamea trivial one. It was left mainly with the role of undertaking banking studies through its technicalcommittee.30

27On the 16th of July 1939 the Minister of Finance, sent an explanatory note included the following: "..The government

hopes to see the day when the majority of the share holders [of the NBE] will be Egyptians and trusts that this willcome in the not too distant future...It would have been possible for the government to Egyptianise the capital of thebank by buying all or part of its shares. But it is evident that such an operation is beyond our financial resources...Withregard to the Egyptianisation of the bank in other directions, ..., We have stipulated that the recruitment of the bankstaff will be limited to Egyptians, except in special case...We have made likewise that the Board should have a majorityof members of Egyptian nationality...Foreign directors shall be replaced by Egyptians whenever a seat becomesvacant...Finally We have provided that the President of the Board shall be an Egyptian." See the translation of thisexplanatory note in NBE (1948), op. cit., pp. 85-88.

28 See NBE (1974), pp. 10-15 & p. 19 and Issawi (1963), op. cit., pp. 249-251.

29 Issawi, (1963), p. cit., p 268.

30Doghaim (1989), op. cit., pp. 46-50.

11

II.2.1 State ownership of banks as a form of financial repression:The financial market in LDCs has always been subject to substantial government intervention. Suchintervention is not always justified by regulatory purposes and/or intentions for correcting marketfailure. In most cases government intervention in LDCs can be considered repressive, as far as thefinancial market is concerned. Governments impose an array of measures that deviate the operationsof the financial system away from the market discipline and result in various distortions.

Financial repression can be explained by different arguments, such as the condition of the financialsystem in the less developed countries after their independence, the impact of the dominatingideologies during the late 1950s and 1960s, and/or anti usury laws, but the main common cause forrepression was financing the budget deficit. Financial repression in Egypt took different forms suchas setting ceilings on interest rates, high reserve requirements, directed credit schemes, interventionin the portfolio composition of banks in addition to extracting revenues from the inflation tax anddirect state ownership of banks.

When Egypt started to modernise its economy in the 1950s, after gaining its political independence,the financial system was comprised of foreign owned commercial banks, with the exception ofBanque Misr and some scattered Egyptian shares in some financial units. The activities of thesebanks were concentrated in short-term trade and commercial credit. There was also geographicalconcentration as most of the banks were established in Cairo and Alexandria with very few branchesin other big urban centres. The financial system was segmented and shallow in the sense that eithersome financial services and instruments did not exist at all in some areas, like insurance services,or they existed but were in an inadequate form, e.g. agricultural credit.

Encouraged by segmentation and shallowness, operating banks behaved in an oligopolistic manner.This was facilitated by the absence of a formal legislator. The National Bank of Egypt, as an actingCentral Bank, was not adequately empowered to exercise the known functions of a well establishedmonetary authority in terms of supervision and regulation.31 Asymmetric information, regarding boththe services of the financial sector and potential borrowers, was a result of such oligopolisticenvironment and a cause for other problems, such as adverse risk selection and the application ofnon-price criteria for the allocation of credit, i.e. depending on the reputation of the borrower,political pressure, etc.32

These characteristics are symptoms of market failure which called for corrective intervention by thegovernment. However the intervention, especially during the 1960s was not necessarily correctiveas it was mainly driven by ideological motives, which left the entire financial system publicly ownedand managed. The repression of the financial sector was one of the components of overallinterventionist policy embraced by the government in the 1960s under the conviction of socialistideas as a remedy for economic problems. According to these ideas, the public sector was considered

31ibid.

32See Killick (1993), op. cit, p. 256.

12

the engine of economic growth. Whereas the private sector was regarded as both economicallyinefficient, in undertaking the large projects of the ambitious development plans, and politicallyunreliable because of its close association with the former regime. Hence there was a series ofEgyptianisation measures, de facto nationalisation, of foreign owned enterprises, including financialintermediaries in late 1950s and the comprehensive nationalisation measures of early 1960s.

Consequently publicly-owned banks comprised the entire financial system until 1975 and despitethe allowance for some foreign and private banks to operate, public banks have been its dominantcomponent afterwards. Although owning financial intermediaries by the state cannot be alwaysconsidered a facet of financial repression, we argue that the way they functioned in Egypt, as in otherLDCs, made them a catalyst for repression and a promoter for its unfavourable effects. In thisenvironment the private sector found it hard to compete for credit, simply because credit isadministratively allocated to the priority projects.

Under Law 40 of 1960 both the NBE and Banque Misr were nationalised by converting their sharesto government bonds33which could be redeemed after a minimum of 12 years. While thenationalisation of the NBE was justified by the fact that it was the State Central Bank, thenationalisation of Banque Misr was mainly because of the government's anxiety to obtain controlover its affiliated companies. However, one year later, all banks were nationalised, under Law no.117 for 1961 which was one of a group of laws nationalising the main economic establishments inthe country to safeguard the creation of a centrally planned economic system as claimed at the time34.

In 1961 the NBE was divided into two banks; one kept the same name and carried on as acommercial bank and the other was called the Central Bank of Egypt. In 1963, a 'PublicOrganisation of Banks' was formed, replacing the League of Commercial Banks, to assist the CBEin controlling and supervising the banking units. To avoid dualism in activities, this organisation waslater abolished and the CBE acquired its supervisory functions in 1964. However the TechnicalCommittee which was established by the organisation was left intact to play a consultative role.

After liquidating some of the small banks and merging others, by the end of 1963 the banking systemconsisted of only five public commercial banks, namely: the NBE, Banque Misr, Banque du Caire,Bank of Alexandria and Port Said Bank; in addition to five specialised banks: an agricultural bank,an industrial bank and three real estate banks. In 1971 several banks were merged: the IndustrialBank into the Bank of Alexandria, Port Said Bank into Banque Misr and the Crédit Hypothécaired'Egypte into the Crédit Foncier Egyptien35. By 1974 there was just four commercial banks and threespecialised banks, in addition to 3 unregistered banks, which were the only banks established duringthis period.36

33 Note that Law 263/1957 determined that distributed profits not to exceed 20% of the nominal value of the shares of

NBE. Under the new Law 40/1960 shares were converted to bonds bearing just 5% interest.

35 See NBE (1974), op. cit., p. 19.

36Note that exemption from registration with the central bank means exemption from banking, credit and exchange rate

13

II.2.2 Sectoral and functional specialisation:Under central planning and the domination of public sector, most of loanable funds mobilised bybanks had to be directed towards financing budget deficits and public sector projects37. However thegovernment found it necessary to determine how this financing should be achieved. First in 1964 asystem of sectoral specialisation among the commercial banks was put forward. Under this systemeach economic sector had to concentrate its transactions within one particular bank of the fivenationalised commercial banks.38

The government believed that by concentrating the operations of a sector with a particular bank, thelatter would gain a deeper knowledge about the requirements of the sector. This would help theplanning institutions and the government in investigating the aspects of strength and weakness in thesectors concerned. It is worth noting too that due to the heavy seasonal financial requirements ofmain crops, such as cotton and rice, their financing was exempted from the sectoral specialisationsystem and all banks participated in covering their credit requirements. The system did not cover theprivate sector as its members were free to select the bank they wished to deal with, regardless of theiractivities.39

After eight years of its establishment in 1964, the government realised that the sectoral system wasnot practical and replaced it with another controversial system known as a functional specialisationsystem in 1971. Under this system, each bank had to serve a specific economic activity. For examplethe NBE specialised in foreign trade in addition to issuing and servicing savings certificates; BanqueMisr specialised in local trade and agricultural crops; and Bank of Alexandria, in which the IndustrialBank was merged, handled all the agricultural and industrial activities. Financing the services sectorwas the responsibility of Banque Du Caire, while the Crédit Foncier Egyptien specialised inconstruction and housing finance.40

This system, like its predecessor, was not based on a theoretical framework or practical guideline;

control laws. The first of the unregistered banks was the Arab African Bank which was established in 1964 in the formof a joint stock company, its dealings were confined to foreign currencies. The second bank was Nasser Social Bankwhich was established in 1971 with the aim of promoting social cooperation and solidarity. Nasser Social Bank isofficially classified as an Islamic bank. One of its important functions was the collection of alms tax and charities toassist poor families by helping them to establishing small projects. The third unregistered bank was the InternationalArab Bank and its dealings were restricted to foreign currencies

38Under the sectoral specialisation system each bank was assigned some sectors to finance. Thus, Banque Misrspecialised in textiles, weaving and light industries; the NBE was responsible for financing land reform projects ,transportation, defence, Suez Canal, Railways and communications; Bank of Alexandria specialised in financingindustrial companies and the Public Organisation of Petroleum; Bank du Caire financed foreign trade, housing andpublic utilities, information and tourism; Bank of Port Said was responsible for financing domestic trade, health andpharmaceutical industry. See Refaie (1989), p. 3.

39 CBE (1984) , p. 284.

40 SEE CBE (1984), op. cit., pp. 286-287, and NBE (1989), pp. 241-242.

14

it was merely an artificial specialisation experiment. Effectively, banks were granted monopolisticpositions in providing credit to economic activities assigned to the respective banks. The bankingsystem consequently suffered from lack of competition and there were calls from the public sectorcompanies to be left free to choose their sources of credit. From the banking system side, there wasalso a demand for equal opportunities, as it was felt that some banks were allocated more profitableactivities than others. As a result this system had to be abolished few years later, as part of theintroduction of the open-door policy.

II. 3 The third phase 1974-1991Partial liberalisation of the banking sector:The period of war economy, 1967-1973, witnessed the cancelling of the second five year plan, aseconomic resources were directed to warfare. This period left the economy with a massive tradedeficit and an immense shortage of financial resources needed for rebuilding what the two wars haddestroyed and restructuring the malfunctioning economy. It was realised that an improvement in thestructure of the banking system and an associated reform of credit policy was essential in order tofind new resources of finance and encourage the private and foreign capital to participate in thedevelopment process. Hence the government started to adopt in 1974 what has become known asthe Infitah or open door policy.41

The Infitah policy which was accompanied with a windfall of external resources stemming from oilexports, workers' remittances, tourism, the Suez Canal revenues and foreign assistance had a radicalimpact on the banking system and structure. Under the Investment law 43 for 1974 and itsamendments by law 32 for 1977, banks were no longer subject to the exclusive Egyptianmanagement rules of the 1960s. Foreign capital was allowed to participate in joint-venturecommercial banks, under the no-less-than 51% national ownership condition.

In an effort to overcome obstacles and eliminate restrictions hindering the banks from achieving thetargets of the open-door policy, law 120 for 1975 was issued. It was promulgated with a view todefining the legal status of the CBE and its supervisory and regulatory powers over the bankingsystem; giving the banking system more freedom in conducting its business; reinforcing the CentralBank's ability to manage monetary policy; freeing bankers from the laws, decrees and restrictionsapplicable to government and public sectors employees; and enabling the nationalised banks tocompete with those established under Law 43 for 1974.

According to law 120 for 1975, the CBE's role was re-enforced by more provisions than those givento it by law 163 for 1957. The Central Bank became an independent legal authority, vested with morepowers for defining monetary and credit policies and supervising implementations in accordancewith the national plan.42 After a series of undisciplined actions by some banks which embarked onrash competition to attract the biggest number of clients, along with the increase and spread of

42 In order to reinforce the CBE independence and its stability Law 120 for 1975 stipulates that the CBE Governor shallnot be dismissed during his term of office and an annual report has to be submitted by him to the People's Assemblywithin 3 months from the end of fiscal year.

15

banking units, it became necessary to amend the two laws 163 for 1957 and 120 for 1975. Thus inMarch 1984 Law 50 for 1984 on the central bank and the banking system was promulgated. The lawestablished explicit and firm rules to prevent exploitation of power by members of board of directorsof banks and gave the Minister of Economy, on the recommendation of the governor of the CBE,the right to veto the appointments of members of these boards. Also under this law no bank maygrant any single client credit facilities exceeding 25% of its paid-up capital and reserves.43

The Infitah laws encouraged the establishment of foreign and joint venture banks. Hence the numberof operating banks increased rapidly from 7 banks in 1974 to 81 banks in 1991. The increase in thenumber of banks also resulted in a rise of the number of branches which reached a total number of1882 branches in 1991.44 However, during this period the structure of the banking system and thegeographic concentration of branches, indicated a highly segmented market and lack of competition.The relatively large branch networks of a few public sector banks allowed them to dominate theprocess of savings mobilisation and other banks targeted mainly big savers.

Financial services during this period were basic with very limited innovation, though joint-venturebanks were in a better position. Lack of innovation and limited instruments can be attributed to threemain factors: lack of competition, the imposition of ceilings on interest rates and intervention incredit allocation. In addition, restrictive measures applied by the CBE regarding the charges forbanking services did not make financial innovation viable or worthwhile. Moreover, the virtualabsence of active money and capital markets, has prevented the operating banks from diversifyingtheir instruments according to risk and maturity.

Thus, lending by public and other banks has been effectively controlled by the CBE. The remainingamount of credit left to the discretion of banks was directed mainly to relatively large enterprises.Lending to small and medium-sized projects suffered from a lack of information and collateral.Further under financial repression, there was a tendency to lend according to non-price criteria wheninterest ceilings are binding. Applied lending criteria in such conditions, included the reputation ofthe client, size of loan and political pressure making large enterprises more likely to obtain credit.

The dominance of public banks in developing countries is identified with familiar problems whichtake two forms: First, internal management problems resulted from lack of incentives, politicalinterference, overstaffing and lack of managerial and banking skills. Second, the general economicenvironment and policy background according to which public banks are assumed to operate.45

Public banks during the study period were found to be protected by an array of regulations andpreferential treatment46, for example: branching by private banks was more restricted than branching

43 This ban did not apply to government agencies, public authorities or state owned companies.

44See CBE, annual report 1990/91, p. 132.

45Fry (1990), p. 15.

46For further details see World Bank (1992), pp. 1-41.

16

of public sector banks and credit ceilings were set in a way that favours public banks47; private bankswere denied the opportunity to provide certain financial service to the public sector companieswithout the acceptance of their assigned public banks and pension funds of the public sector had tobe deposited with public banks. In addition, the design of credit ceilings favored public sector banksas they were imposed on activities mainly undertaken by private banks, like credit to the commercialsector. Payment of interest on current accounts was prohibited, benefiting public sector banks in twoways. First, they did not bear any costs for keeping such deposits which mainly belonged to thepublic sector enterprises. Second, private banks were not able to compete for such accounts byoffering higher interest rates to depositors.

Public banks in Egypt like in other LDCs were more prone to government interference in credit andplanning decisions than private banks. Incentives to maximise profits, or even to minimise losses,barely existed. Moreover public banks lacked managerial autonomy and like the rest of governmentsectors units in Egypt they had no effective power regarding staffing decisions. The conditions ofgovernment owned financial intermediaries in India which, more or less, describe prevailingconditions else where were summarised as follows:48 (i) low resource mobilisation; (ii) lowprofitability; (iii) low capitalization ratios and insolvency; (iv) complicated bureaucratic proceduresfor loan processing and operating inefficiency; (v) allocation of resources on the basis of non-economic criteria; (vi) reduced autonomy; (vii) poor quality of personnel, overstaffing and weakmanagement.49

Nevertheless we argue that the remedy of such problems does not lie merely in changing theownership of banks. It is established that it is not the type of ownership per se which determinewhether a bank would be an aspect of repression, it is rather the mechanism according to whichpublic banks operate that emphasise the negative effects of repression and hinders competition.50

Evidence from some OECD countries such as Canada, Sweden and the Netherlands, show that stateownership can be compatible with low operating costs and high efficiency of intermediation. Thisis primarily attributed to market competition and contestability.

II.4. The fourth phase 1991-Repressed liberalisation of the banking sector:The Iraqi invasion of Kuwait in 1990, resulted in massive returns of Egyptian workers from theconflict zone, losses of workers remittances, a decline of tourist receipts and Suez Canal dues anda worsening of investment climate. However, Egypt received significant financial assistance fromthe Gulf states and the USA which cancelled US$ 13 billion of its total external debt of US$ 51billion. Egypt was also granted, in 1991, a debt/debt service relief by the Paris Club creditorsequivalent to 50% of the outstanding debt over three phases through mid-1994 conditional upon 47 When the CBE in the late 1980s adopted a strict policy to control the number of branches, the number of branchesper public bank was significantly higher than that of other banks.48Cited in Fry (1991) from a study by Morris (1985) on the Indian financial system.

49On bank efficiency losses and rules for restructuring banks in transitional economies see Andrew Sheng (1990).

50Killick (1993), p. 272.

17

implementing an IMF agreement.51

This financial assistance, accompanied with higher oil export prices and a decrease in main importprices, especially foodstuffs, offset the losses incurred as a result of the Gulf crisis. The balance ofpayments experienced a surplus of around US$ 0.9 billion in 1990/91, the debt forgiveness andrescheduling reduced the debt service from 46% of exports to approximately 16.5% and permittedan increase in the net international reserves from only US$ 1.7 billion to US$ 6.1 billion.52

These developments improved the creditworthiness of the Egyptian economy and improved itscapacity to embark on a comprehensive Economic Reform and Structural Adjustment Programme(ERSAP) in 1991. This reform programme was supported by a stand-by arrangement from the IMFand a structural adjustment support from the World Bank, in addition to the bilateral debtforgiveness/debt service relief of the Paris Club. The primary objective of ERSAP is summarisedby the IMF (1991) as

"to create, over the medium term, a decentralized market based, outward-orientedeconomy where private sector activity will be encouraged by a free, competitive, andstable environment with autonomy from government intervention. For this purpose,controls on economic activity and investment are to be dismantled and primaryreliance placed on market forces for resource allocation."53

This reform programme has involved the financial sector in three main ways.54 First, channellingloanable funds into budget deficit, in a way that reduces the reliance on inflationary finance until thedeficit itself can be controlled. Second, advancing finance to the existent and emerging private sectoractivities. Third, the financial sector in its own right has been subject to an array of liberalisation andinstitutional restructuring measures included in the programme. Hence the financial sector isconsidered as being in the forefront of the reform programme as an as an 'agent of change'.55

The reform of the banking sector during the 1990s took three main forms. First, liberalisation offinancial variables, by mechanical reversal of financial repression measures adopted since early1960s. Hence, interest rates ceilings were eliminated, the required reserve ratio were reduced,liquidity ratio decreased, the government minimised its intervention in credit allocation and theportfolio composition of banks. Figure (1) below shows that real interest rates have turned positivein 1991 as a result of financial liberalisation through the lifting of ceilings on interest rates. However,the relatively high real interest rates which reached 7.9% in 1999, indicate, inter alia, lack ofcompetition between the operating banks.

51IMF (1991), pp. 7-8 and World Bank (1992), p. xxv.

52ibid.

53 IMF (1991), p. 8 and for the details ERSAP, see World Bank (1991) and (1992), and IMF (1991) and (1992).

54Harvey and Jenkins (1994), p. 1.

55 Wijnbergen (1993), p. 9.

18

Source: Calculated from IMF, International Financial Statistics, various issues.

Second, as part of the banking reform program, one of the main priorities of the authorities wasimproving the regulatory framework. New prudential measures were introduced in 1991 for capitaladequacy and asset classification and provisioning after setting bank liquidity requirements fordomestic and foreign currencies in 1990. Other regulatory measures followed: internationalconcentration limits were introduced in 1992, and domestic concentration limits in 1993. By 1996,the regulatory framework was in compliance with international norms, and banks were required toproduce periodical financial statements based on international accounting standards since 1997.56

Third, the banking sector was included in the privatization program which took the form of theprivatization of the public sector shares in joint venture banks, and the public banks. The 23 jointventure banks were mainly established during the free entry of the 1980s, most of which weremajority owned by the public sector. During the 1990s there was a reduction of the public sectorshares in joint venture banks.57

56 IMF (1998) pp. 58-60 and for a discussion of prudential regulation in Egypt see Bahaa Eldin and Mohieldin(1998), pp. 111-135.57 Sale of public sector shares in joint venture banks took several forms, the sale of shares on a stock market through acombination of a local IPOs, and International Global Depository Receipts (GDRs), increases in capitalization of existingprivate sector owners, and private placements to local and foreign partners.

Figure (1)Real Interest Rates (1961-1999)

-15%

-10%

-5%

0%

5%

10%

15%

1961

1963

1965

1967

1969

1971

1973

1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

19

Privatization of joint venture banks has been facilitated by several regulatory measures. The fourpublic banks were requested in 1994, to reduce their shares in joint venture banks to less than 51%,and in early 1996 they were requested to further reduce their shares to a maximum of 20%. Moreover, majority foreign ownership was then permitted in joint venture banks through passingLaw 97 of 1996. Privatizing the public sector shares in joint venture banks is expected to lessen thecommercial interdependence between different banks. No major activity in the privatization of jointventure banks occurred until early 1996 and despite, several divestiture of public shares in jointventure banks, total privatization of the public banks’ shares in joint venture banks has not beencompleted.

As shown in table (1) below, as of March 2000, out of the 23 joint venture banks, seven arestill majority owned, by the public sector, of which two are majority-owned by the four public banks,three are majority-owned by other public entities,58 and two are jointly owned by both the publicbanks and other public sector entities. The public sector has less than 20 percent in eight other jointventure banks.

58 These public entities include Public Insurance Companies, NIB and Public Specialized Banks and Authorities.

20

Table (1)Capital Structure of Joint Venture Banks as of March 2000

Joint Venture Bank NBE BanqueMisr

Banque duCaire

Bank ofAlexandria

TotalPublicBanks

OtherPublic

Entities*

TotalPublicShare

Alexandria Commercial & Maritime Bank 0.0 52.2 52.2Banque du Caire et de Paris 22.0 22.0 22.0Cairo Barclays International Bank 40.0 40.0 40.0Cairo Far East Bank 29.0 29.0 20.0 49.0Commercial Arab Bank 9.8 9.8 9.8Commercial International Bank (CIB) 19.6 19.6 19.6Credit Internationale d’Egypte 19.5 19.5 19.5Egypt Arab African Bank 0.0 18.2 18.2Egyptian American Bank (EAB) 35.3 35.3 35.3Egyptian Gulf Bank 0.0 0.0 24.4 24.4Egyptian Saudi Finance Bank 7.7 8.5 7.4 23.6 12.7 36.3Trade and Development Bank 17.0 16.0 16.5 9.7 59.2 27.2 86.5Export Development Bank 11.5 11.5 11.5 10.1 44.5 40.0 84.5Housing & Development Bank 0.0 62.5 62.5Islamic Bank for Investment & Development 20.0 20.0 20.0 20.0 80.0 80.0MIBank 24.8 24.8 4.0 28.8Misr America International Bank 32.8 32.8 67.2 100.0Misr Exterior Bank 19.5 19.5 19.5Misr Iran Development Bank 37.5 37.5 37.5 75.0Misr Romanian Bank 33.0 33.0 33.0National Bank for Development 0.0 11.9 11.9National Societe General Bank 18.0 18.0 18.0Suez Canal Bank 4.6 0.1 4.7 12.3 17.0

Source: Nasr (2000), p. 19.

Further, the parliament in June 1998 passed Law 155, which allows for private sector participationin the ownership of the public banks. However, no action has been taken regarding the privatisationof any of the public banks. This has been explained by giving the priority for the sale of publicownership in the joint venture banks and then the privatisation of the public banks may beconsidered in the future.59

After almost three decades of partial liberalisation of the banking system and a decade of relativelymore comprehensive financial liberalisation within the economic reform program, there has been anotable increase of the number of operating banks and a corresponding increase in the number ofbranches, from 30 banks and 530 branches in 1975 to 62 banks and 2444 branches in 1999. However, the banking density60 at 0.37 in 1999 is still quite modest, which is not better than bankingdensity in 1980 which was 0.34. Given the high concentration of branches in particular parts inGreater Cairo and Alexandria, the distribution of banking services, at such low average bankingdensity cannot be considered as sufficient. Figure (2) shows that after some improvement in thebanking density, not adjusted for distribution, during the 1970s and 1980s, it remained almostconstant during the 1990s. 59 Reuters in the 19th of September 1999 reported that President Mubarak in an interview with October, an Egyptianweekly magazine, declared that “Maybe, at some point in time, we will think about offering a stake but that is notbeing considered at this point in time”.60Bank density = (Number of branches *10000) /Total population. A density of 1 is high, .5-1 moderate and below .5

is low, see Cameron (1967), p. 296.

21

Source: Central Bank of Egypt, Annual Report, various issues.

Figure (2)Bank Density

during the Period (1970-1999)

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40

1970

1975

1980

1984

1986

1989

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

22

Moreover branches of the four public sector commercial banks comprise 37.6% of total branchesin 1999 and witnessed an increase of 2.4 percentage points compared with 1991 at the start offinancial liberalisation. Figure (3) reveals an interesting development of the public sector banksduring the 1990s, as far as the growth of their market share is concerned. While there is a negativegrowth rate of the public banks’ share in the total loans, deposits and assets of commercial banks,their branching growth rate continued to grow steadily.

During the 1993-1999 period, public sector banks share in total assets of commercial banks declinedfrom 74.6% to 63.1%; their share in total deposits declined from 75.5% to 66.1%; and their sharein total loans decreased to only 59% from a high level of 92.6 at the beginning of the period.

Source: Calculated from the CBE Economic Bulletin and Annual Reports.

Figure (3)Growth of Public Sector Banks’ Share in Total Assets, Deposits and Loans,

and Growth of Branches during the Period (1994-1999)

-20

-15

-10

-5

0

5

10

15

1994 1995 1996 1997 1999

Total Assets Total Deposits Total Loans Branches

%

23

III. A note on the SCP paradigm in the Egyptian context:

Benefiting from the advances in the field of industrial organisation, many empirical studies havesought to analyse the relationship between market structure and various aspects of bank conduct andperformance as suggested by the structure-conduct-performance paradigm (hereafter SCP).61

According to Caves (1967)62, in his analysis of the American industry, market structure determinesthe behaviour of firms and that behaviour in turn determines the performance of the industry at large.Moreover, It has been argued that fewer and larger firms, i.e. higher concentration, are more likelyto engage in anti-competitive conduct. According to this paradigm causation runs from the structureof the banking market to performance, through the behaviour or conduct of the banking units, in anessentially deterministic manner.63

Empirical tests of the SCP paradigm in banking industry have often found that the structure of themarket, using market concentration as a proxy, has influenced both conduct and performance. Gilbert(1984) concludes in his comprehensive survey that the majority of the evidence supports the idea thatbanking industry structure influences bank performance.

It is also established in the industrial organisation literature that contestable markets and potentialfirms' freedom of entry promote efficiency, encourage innovation and give highly favourable welfareoutcomes.64 For a market to be contestable there should not be any significant entry barriers. In acontestable environment the only way for incumbent firms to prevent additional entry is to give noincentive for potential entrants to do so. This can be achieved by efficient pricing and allocation ofproduction among incumbent firms to eliminate significant excess profits. However in practice thereare different barriers to entry facing potential firms which prevent them from joining the market evenin the presence of high excess profits. Large economies of scale and high sunk costs, in addition toother entry costs are examples of such barriers.65 But in the case of banking, government regulationsthrough permits and licenses are far more important than these other barriers. As indicated above, we argue that the Egyptian banking system has suffered from two mainproblems. The first was the application of financial repression measures, such as interest rate ceilingsand directed credit which virtually prevented competition among incumbent banks. The second wasthe use of restrictive regulations that prevented new entry and made the incumbent banks far frombeing contestable.

The banking system, after a series of Egyptianisation and nationalisation measures in the 1950s and1960s, was left with four publicly owned commercial banks and five specialised banks. The marketwas highly concentrated. Competition was limited further by the application of sectoral and, then, 61 For a comprehensive survey of the empirical studies see Gilbert (1984).62 Cited in Sawyer (1981), p.154.63 ibid.64For an analysis of contestable markets see Baumol, Panzar and Willig (1982).

65 On entry barriers see Bain (1956) and Tirole (1989), chapter (8).

24

functional specialisation which made the system a sectoral based mono-bank one. The introductionof Infitah policy of the 1970s and the financial liberalisation measures of the 1990s resulted in anincrease in the number of banking units without a significant decrease in market concentration. Thedomination of public banks in a highly concentrated market resulted in frail competition and limitedinnovation.

The main reasons behind restrictive regulations regarding new entry can be summarised by sixfactor: first, concerns about "cream skimming" by private and foreign banks. Second, fear of acquiring dominant positions by foreign banks in the domestic market. Third, concern about hit andrun activities by foreign banks. Fourth, protecting the interest of the incumbent banks, especially thepublic ones. Fifth, concerns regarding extensive allocation of mobilised domestic savings abroad.Sixth, regulatory concerns based on a claim that foreign and private banks may require particularregulatory capacity.66

It was not only the restricted entry that made the Egyptian banking market suffer from lack ofcompetition and contestability, restricted exit was also responsible. Indeed, whereas it is importantto remove barriers to entry, it is also crucial to maintain a reliable exit mechanism to improve theefficiency and the soundness of the Egyptian banking. In Egypt banks are not allowed to fail. Thispolicy was given effect neither through prudential policy, nor measures that enhance the efficiencyof banks. Instead, weak banks were allowed to continue in business using support from the rest ofthe banking system and public money. This was ‘justified’ by the fear of a public misunderstandingthat the failure of bank may imply that others will follow in the future, made the banking systemadopt a form of collective self-preservation.

According to this approach inefficient banks were left to operate through support from the bankingsystem, while adequate measures like restructuring, merging or liquidation were not applied. Thispolicy encouraged inefficient banks to continue their violation of credit standards by indulging inhigh risk lending and bidding for deposits.67 In addition, because all banks are supported by animplicit rescue mechanism in which bad banks are cross subsidised by good ones, banks' clientscannot distinguish between an efficient bank and an inefficient one.68

Three different types of operating ratios are frequently used as indicators of the performance of banksand their efficiency, namely rate of return on assets (ROA); rate of return on income, (ROI); and rateof return on equity (ROE). ROA relates net income to total assets; ROI relates net revenues to the

66 For further discussion of these arguments see Bahaa Eldin and Mohieldin (1998), pp. 130-131.67 Even under the application of interest ceilings, a bank like the Bank of Credit and Commerce International, which was

registered in Egypt as a joint-venture bank was paying a higher interest rate on deposits, than the rest of banks, by 0.5-1percentage points, till its collapse in 1991.

68 Under the implicit deposit insurance scheme, the depositors of the collapsed BCCI were protected by transferring theiraccounts to Banque Misr at their full nominal values at the date of the BCCI's collapse. Registered banks were askedto contribute 0.5% of their deposits towards the funding of this operation in addition to a one billion pound interest-freeloan to Banque Misr to accept the 'transfer' of BCCI accounts to its branches. See Bahaa Eldin and Mohieldin (1998),p. 128.

25

ROAit = α + α1MCRt + α2MSit + α3CAit + α4TAit + α5LAit

+ α6DTit + α7OEAit + α8MDGRt +α9D (1)

total of interest and non-interest income i.e. gross income; ROE is the ratio of net income to averageequity.69 Three issues are worth emphasising: first, these ratios should be considered only as roughindicators of efficiency as these ratios are significantly affected by differences in capital structure,accounting methods used and product mix, both across countries and banking units.70 Second, bankdata in developing economies should be treated with caution. Until the adoption of the EgyptianAccounting Standards, which are compatible with the international ones, in 1997, different banksused different accounting practices regarding their valuation of the quality of assets and depreciation.Accordingly their provisions and reserves may differ as well. Some banks, in developed anddeveloping countries may indulge in some form of window dressing. Such practice distorts thequality of information and produces misleading indicators.

V. Model specifications, data and preliminary results:

We analyse the relationship between a measure of performance (ROA), as a dependent variable, anda proxy for market structure (MCR) and other independent variables that include both bank andoverall market specific variables. We formulate a cross-sectional profit equation similar to thoseused in previous studies that followed Weiss (1974) and Smirlock (1985). More recently, the sameapproach has been used by a study of the effects of financial liberalisation on market structure inTurkey by Denizer (1997) and another one on the relationship between market structure andperformance in the Jordanian banking system by Al-Karasneh, Cadle and Ford (1997). Theestimated equation takes the following form:

MCR as a proxy for market concentration,71 is calculated as the sum of shares of the 5 leadingEgyptian banks in total deposits. Here, the use of a proxy is justified by the fact that industrialorganisation literature does not specify an exact critical minimum of number of firms or their sizedistribution to exercise market power. The market share (MS) variable which is calculated as bankdeposits divided by total market deposits is again a proxy for firm-specific effects. Influenced by theSCP literature, the equation also includes a number of control variables, from the assets andliabilities sides of the banks’ balance sheets, to internalize the possible effects of factors such as risk,costs and demand that influence our measure of profitability (ROA) as a dependent variable.

The equation includes a capital-asset ratio (CA) which account for the bank specific risk exposures;total assets (TA) gives an indication of the size of individual banks and possible economies of scaleand product diversification associated with it; the ratio of total loans to total assets of the individualbanks (LA), shows the extent of bank’s risks, as higher ratios of loans to total assets reveals theaggression of lending by the bank to increase profits. The amount of demand deposits relative to total

69See Brealy and Myers (1991), chapter (27) for an analysis of financial performance.

70See World Bank (1992), pp. 25-26 and Vittas (1991).

71 We adopt the same approach used in Denizer (1997), pp. 18-22 in the calculation of the reported ratios.

26

deposits (DT) is also included to give an indication of relative cost of funds. Operating expenses(OEA) as a proportion of total assets is included as a cost factor that reduces bank profits. Theequation take into account the impact of market demand by including the market deposits growthrate (MDGR). Finally, a dummy variable D is included to differentiate between different types ofbank ownership; for private banks D equals 1, and for public banks D equals 0.

The data have been specifically obtained from the individual balance sheets of 47 commercial banks,out of a total of 59 commercial banks. The rest of the commercial banks are not included eitherbecause of no response to a series of data requests during a six months period of data collection, orbecause of insufficient and/or missing data for parts of the period under study. The sample represents79.6% of the total number of operating commercial banks, but more importantly comprises 94.6%of the total assets of the commercial banks in 1998. The sample data cover the period between 1980and 1998. The sample contains the required information to calculate the 10 variables included inequation (1). The data which include 893 observations have been pooled, to allow for cross-sectiontime series analysis.

Equation (1) is estimated using OLS in three steps. In the first step, equation (1) is estimated withboth market structure (MCR) and market share (MS) variables. Here two competing hypotheses aretested: the first one is the traditional SCP that establishes a positive relationship between MCR andprofits, the second is the efficient structure hypothesis that suggests a direct positive relationshipbetween profits and market share attributed to firm specific efficiency, which is established if α2>0and α1=0. But if α1>0 and α2=0. This will mean that market share (MS) does not affect ROA andhigher profitability reflects only market concentration (MCR). In the second step, equation (1) isestimated to test the traditional SCP hypothesis, by eliminating market share from the equation.Finally, in the third step the effect of market share on profits is tested by estimating equation (1)without market structure variable.

Using White test for heteroskedasticity shows the absence of such problem from the pooled data. Theexplanatory power of the three equations is acceptable, bearing in mind the cross-sectional natureof the data-set. Table (2) summarises the results of the three estimated versions of equation (1).

27

Table (2)Dependent variable is Return on Assets, ROA

1. 2. 3.Variable

Coefficient t-Statistic Coefficient t-Statistic Coefficient t-Statistic

C -1.283403 -0.988285 -1.468596 -1.131045 -0.158093 -0.573765

MCR4 0.357176 0.886740 0.414205 1.028428

MS 0.116897 2.155732 0.120055 2.219029

CA 0.296550 3.331066 0.165645 2.539202 0.299192 3.363033

TA -0.230727 -7.806664 -0.213456 -7.487802 -0.232069 -7.863336

LA 0.026701 0.394647 0.050821 0.760065 0.018955 0.282549

DT -0.317475 -3.858509 -0.354017 -4.387925 -0.321772 -3.918019

OEA -0.074120 -0.912405 -0.069794 -0.857646 -0.074712 -0.919844

MDGR -0.284127 -2.068307 -0.248824 -1.820578 -0.288012 -2.097908

D 0.509274 4.272696 0.421639 3.755350 0.507808 4.261323

R-squared 0.264835 0.260966 0.264181

Adjusted R-squared 0.257342 0.254278 0.257522

S.E. of regression 0.820128 0.821818 0.820029

F-statistic 35.34352 39.01958 39.67277

The coefficient on the market structure (MCR) variable is statistically insignificant at 5% level inall equations, while the coefficient on the (MS) is positive and significant at the same level. It canbe argued that in effect MCR and MS explain the same phenomenon72, indicating a high collinearitybetween them and result in one variable, that is MS in our case, becomes statistically significant atthe expense of the other. But those who support the efficient structure hypothesis will reject thisargument, claiming that the significance of MS, as a proxy for firm-specific effects, is due to superiorefficiency, not market power as the traditional SCP school would see it. As far as the impact of the control variables73 is concerned we find that the coefficient of (CA) the 72 Al-Karasneh, Cadle and Ford (1997), p. 13.73 Note that Loan to assets ratio and operating expenses are statistically insignificant.

28

capital asset ratio is significant and positive, indicating that the highly capitalised banks areconservative and hold less risky assets. The significant and negative sign of the coefficient of sizevariable (TA) indicate lack of exploitation of economies of scale or rather an experience ofdiseconomies of scale, and imply that larger banks, which are the public ones, do not necessarilyrealise, inter alia, benefits of product diversification opportunities.

The significant and negative coefficient of the cost of funding variable (DT)74 is rather a surprisingfinding. But given that the banks that enjoyed higher demand deposits to total deposits ratios duringthe study period, i.e. public banks, were not among the highest performing, the surprise may be less.The coefficient of market deposits growth rates (MDGR) is significant but negative which indicatesthat banks did not benefit from the growing demand for deposits in a way to increase their profits.Although it can be argued that markets with high growth rates are more likely to increase the depositbase of banks but it should not be taken for granted that this would necessarily lead to higher profitsof the concerned banks. This obviously depends on the ability of banks to convert deposits, asliabilities, into performing assets under the prevailing macroeconomic and sectoral conditions.75

Finally, the significance of the coefficient for bank ownership and its positive sign emphasise whathave been implied above through suggesting that private and foreign banks are more profitable thanthe public ones. This finding is supported by the average ROA figures for public banks comparedwith the non-public ones during the last two years. The average of the four public banks was 0.50%,while the average of four leading commercial JVs76 was 1.82%.

74 A typical public sector bank like the NBE had 27.1% of total deposits in a form of current account in 1991 while theratio in the Delta Bank, a joint-venture bank, was only 9.7%.74

75 Denizer (1997), pp. 21-22.76 The four banks are CIB, MIBank, EAB and EBB.

29

V. Concluding remarks and policy implications:In this chapter we have shown that the development of the banking system in Egypt can be dividedinto four phases. During the first phase 1856-1956 the banking system was highly deregulated anddominated with foreign commercial banks that offered short term credit mainly for the cultivationand trade of cotton. The close relationship between the operating banks and the cotton market madethem highly vulnerable to the severe and sudden fluctuations in its trade due to external shocks, e.g.the American Civil War, the great depression, the two World Wars,.. etc. The credit market waslargely unstable during this phase with large numbers of banks collapsing and the existence of hitand run banking.

This phase witnessed also three important events as far as banking development is concerned, firstthe evolution of central banking since 1898 with the issuance of banknotes by the NBE which wasfollowed by several decrees enforcing its functions as banker and advisor for the government and,to some extent, bank of banks. The second was the establishment of specialised credit institutionsin 1902 and then in 1931, to provide the agricultural sector with moderately priced credit andcompeting with commercial banks and money lenders. The third was the emergence of Banque Misrin 1920 as the first Egyptian-owned bank. The bank was a universal one but was more concernedwith the establishment of a diversified industrial base.

The second phase that covers the period 1957-74 is characterised by the Egyptianisation of foreignbanks in 1957 as a result of the Suez Crisis of 1956, and then the entire nationalisation of thebanking system in 1961 which initiated the adoption of central planning. As a result of merging andliquidations of banks the number of operating banks declined from 27 banks in 1961 to ten banks.The functions of the central bank were enforced by the banking law of 1957 and the CBE wasestablished as an independent institution in 1961. This period also observed unsuccessful sectoraland functional specialisation policies which had to be abolished.

In the third phase 1974-1991, the banking system was partly liberalised according to the Infitahpolicy. Foreign and private banks were allowed into the banking system and the number of banksincreased substantially from 7 in 1974 to 81 banks in 1991. Despite this increase which was reflectedin a rise of branches, the banking density remained modest and competition was frail. The financialreform program, which started in 1991 with extensive coordination with the IMF and the WorldBank, marked the end of the third phase and the start of the fourth phase.

We have shown that the Egyptian banking system during the 1980s and 1990s, despite the partialliberalisation measures of the Infitah and the more comprehensive financial reform measures underERSAP suffered from various problems. Entry barriers, absence of an adequate exit mechanism, andpreferential treatment of the public sector banks are the main difficulties that impede thedevelopment of the banking system. The emphasis of the financial reform program, in its early yearsof the 1990s, was primarily on the liberalisation of some of the financial variables and theintroduction of prudential measures compatible with the international reform. Very recently,attention is being given to the development of the structure of the banking system and improvingcompetition between the banking units.

30

We have tested the relevance of the structure-conduct-performance paradigm in the Egyptianbanking context. The ‘traditional’ SCP model has been investigated using pooled cross-section timeseries data specfically collected 47 commercial banks, out of a total of 59 commercial banks. Whilethe coefficient on the market share is significant and positive we found that the concentrationvariable, i.e. the market structure variable is not significant. The view of the traditional SCP schoolcan be used to argue that market power enjoyed by the dominating and protected public banks, thesignificant market share variable explains the impact of the structure on bank performance.According to our analysis of the evolution of the bank structure in Egypt, there is very little that canbe said about bank specific superior efficiency in gaining this market share.

According to the analysis we found that highly capitalised banks conservative and hold less riskyassets and larger banks in Egypt, which are the public ones, do not necessarily realise, inter alia,benefits of product diversification opportunities. Moreover, it does not seem from the analysis thatpublic banks benefited as they should from their relatively low cost of funding as they were lessprofitable than foreign and private banks.

The results of the estimated model should be used with care as far as their policy implications areconcerned. The deterministic way in explaining the relationship between market structure andperformance, that the SCP paradigm adopts, is an aspect of straightforwardness and weakness of theparadigm at the same time. The paradigm does not give sufficient consideration for rivalry andcompetition between firms and deals with them as passive economic agents, or a bridge throughwhich structure influences the performance77. The model does not also give an explicit role ofdifferential efficiency of firms on profits, and concentrates on the role of effective collusion betweenthe leading firms.78

Further, any drift from profit maximisation influences the results of the paradigm. Profit as anindication of bank performance is subject to two important challenges in the Egyptian bankingcontext. First, during most of the study period the Egyptian banking system was subject to financialrepression measures that affected the profit maximisation behaviour. Second, given the nature ofdecision making process in the dominating public banks that is subject to ‘external’ on non-economic grounds, the assumption that they are always pure profit maximisers may be ratherunrealistic. Considering such factors can help in better understanding of the estimated model and itspolicy implications. For example, when we analyse the concentration of public banks it should beviewed differently from the concentration of capitalist firms, and when we analyse market share wehave to identify the segment of the market or the relevant market of public banks.

We have shown that the Egyptian banking system during the 1980s and 1990s faced a mixture ofboth the domination of public banks and the high degree of concentration which discouragescompetition on one hand, and the heavy barriers to entry which makes incumbent banksincontestable on the other hand. However, recent financial liberalization measures eliminated interestrate controls, eased the entry of new financial intermediaries, and allowed new types of instruments.Regulatory impediments were relaxed and the Egyptian banking system become more integrated 77 Sawyer (1981), p. 158.78 Gilbert (1984), p. 629.

31

with the world market.

Further reform measures are required at two levels: first, at the banking units level by improvingtheir management, granting them adequate autonomy in decision making, and encouraging trainingschemes and the acquisition of essential resources for efficient banking. Second at the bankingsystem level, by encouraging competition and contestability, within an adequate regulatoryenvironment, removing excessive barriers to entry and replacing them with an objective entry criteriaand establishing a reliable exit mechanism accompanied with an efficient deposit protection scheme.Finally, it is worth referring here to the quotation at the very beginning of this chapter, whichemphasises the role of history and historical accidents on the structure of industry. The evolution ofthe structure of the Egyptian banking ‘industry’ supports this argument.

32

Bibliography

Ali, A. (1957), Banking in the Middle East, IMF Staff Papers, Vol. VI, pp. 51-79.

Al-Karasneh, I, P. Cadle and J. Ford (1997), Market Structure, Concentration and Performance:Jordanian Banking System, 1980-1993, Department of Economics Discussion Paper 97-12, TheUniversity of Birmingham.

Bahaa Eldin, M., Mohieldin, M. (1998), Financial Development in Emerging Markets: The EgyptianExperience, Cairo: The Egyptian Center for Economic Studies.

Bain, J. (1956), Barriers to New Competition, Cambridge, Mass: Harvard University Press.

Baumol, W., C. Panzar and R. Willig (1982), Contestable Markets and the Theory for IndustryStructure, New York: Harcourt Brace Jovanovich.

Brealey, R. and S. Myers (1991), Principles of Corporate Finance, McGraw-Hill, London.

Cameron, R. (1967) (ed.), Banking in the Early Stages of Industrialization, New York: OxfordUniversity Press.

Central Bank of Egypt, Economic Review, various issues.

Davis, E. (1983), Challenging Colonialism, Princeton University Press, New Jersey.

Deeb, M. (1976), Bank Misr and the Emergence of Local Bourgeoisie, in the Middle EasternEconomy, edited by E. Kedouri, Frank Cass and Co. LTD. London.

Denizer, C. (1997), The Effects of Financial Liberalization and New Bank Entry on Market Structureand Competition in Turkey, Policy Research Working Papers, no: 1839, The World Bank

Doghaim, A. (1989), Economics of Banking, Cairo: Madbouly, [in Arabic]

El-Refaie, F. (1998), Financial Intermediation: the Efficiency of the Egyptian Banking System, inFinancial Development in Emerging Markets: the Egyptian Experience, edited by El-Erian, M.,Mohieldin, M., Cairo: The Egyptian Center for Economic Studies and San Francisco: TheInternational Center for Economic Growth.

Fry, M. (1990), ‘Financial Sector and Government Intervention’, International Finance Group,University of Birmingham, IFGWP-90-07.

Gilbert, R. A. (1984), Bank Market Structure and Competition: A Survey, Journal of Money andBanking, Vol. 16, No. 4, pp. 617-60.

33

Hansen, B. (1991), The Political Economy of Poverty, Equity and Growth: Egypt and Turkey,Washington: Oxford University Press.

Harvey, C. and Jenkins, C. (1994), ‘The Role of the Financial Sector in Structural Adjustment inSouth Africa, Zambia and Zimbabwe’, Mimeo, IDS, Sussex University.

IMF (1991), ‘Egypt’, Staff Report for the 1991 Article IV Consultation and Request for Stand-byArrangement, IMF, Washington.

IMF (1992), ‘Arab Republic of Egypt- Recent Economic Developments’.

IMF, (1998), Egypt: Beyond Stabilisation, Toward a Dynamic Market Economy, IMF OccasionalPaper, No.163.

Issawi, C. (1963), Egypt in Revolution, Oxford: Oxford University Press.

Killick, T. (1993), The Adaptive Economy: Adjustment Policies in Small, Low-Income Countries,Washington D. C.: The World Bank.

Mabro, R. (1974), The Egyptian Political Economy 1952-1972, Clarendon Press, Oxford.

Mohieldin, M. (1995), Causes, Measures and Impact of State Intervention in the Financial Sector:the Egyptian Example, Working paper 9507, Economic Research Forum for Arab Countries, Iranand Turkey.

Nasr, S. (2000), Privatizing and Restructuring the Public Sector in Egypt, mimeo, May.

NBE, Economic Bulletin, virous issues

Owen, R. (1981), The Middle East in the World Economy 1800-1914, London: Methuen.

Refaie, El. F. (1989), The Performance of the Banking Sector in Egypt During the 1980s: SomeBasic Issues, Central Bank of Egypt, Mimeo.

Roe, Alan R. (1998), The Egyptian Banking System: Liberalization, Competition and Privatization,the Egyptian Center for Economic Studies, Working paper No. 28, June, Cairo.

Sawyer, M. (1981), The Economics of Industries and Firms Theories, Evidence and Policy, London:Croom Helm, Second Edition.

Shafi’i, Z. (1985), An Introduction to Money and Banking, Cairo: Dar-Al-Nahdah Al-Arabiah [inArabic].

Smirlock, M. (1985), Evidence on the Non Relationship Between Concentration and Profitabilityin Banking, Journal of Money, Credit and Banking, vol 17, pp. 69-83.

34

Tirole, J. (1989), The Theory of Industrial Organization, MA: The MIT Press.

Vatikiotis, P. J. (1991), The History of Modern Egypt, London: Weidenfeld and Nicolson.

Vittas, D. (1991), ‘Measuring Commercial Bank Efficiency: Use and Misuse of Bank OperatingRatios’, World Bank, PRE Working Papers, WPS 806.

Weiss, L. (1974), The Concentration-Profits Relationship and Antitrust, in H. Goldschmid, M. Mannand J. F. Weston (eds.), Industrial Concentration: The New Learning, Boston: Brown andCompany, 184-233.

Wijnbergen, van S. (1983), ‘Credit policy, Inflation and Growth in Financially repressed Economy’,Journal of Development Economics, vol. 13, pp. 45-65.

World Bank (1991), ‘Egypt: Alleviating Poverty during Structural Adjustment’, Washington: WorldBank.

World Bank (1992), Egypt: Financial Policy for Adjustment and Growth, Report No. 10790-EGT,Washington: World Bank.