Competition in Malaysian Islamic banking

(1)

1

UNIVERSITY OF AMSTERDAM

FACULTY OF ECONOMICS AND BUSINESS

Competition in Malaysian Islamic banking

Master thesis

Islamic banking is quite different from conventional banking since it has to adhere to Sharia law. But does this also affect their profitability and competitiveness? This thesis investigates the Islamic banking sector in Malaysia, since Malaysia has the biggest Islamic banking and finance sector in the world. The Boone Indicator and the HHI were used to measure the competition and concentration of the market in Malaysia. The HHI measures show that the Islamic banking sector in Malaysia is moderately concentrated, which may decrease competition. But, this HHI might be skewed, as a result of government intervention caused by subsidies. The Boone indicator shows that Islamic banks are quite competitive based on their ROE, however the difference with conventional banks was statistically insignificant.

Keywords: Islamic banking, Malaysia, competition, market concentration, HHI, Boone Indicator.

Student: Randa Ibrahim Student ID: 10322817

Supervisor: Barbara Baarsma

Supervisor Rabobank: Raphie Hayat

e-mail address: randa.ibrahim@student.uva.nl Date: June 30, 2018

(2)

2 Statement of Originality

This document is written by Student Randa Ibrahim, who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document are original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

(3)

3

Preface

This thesis combines my interest in economics and competition, with my personal interest in Islamic culture. After writing a bachelor thesis about the Islamic mortgages in the Netherlands for my Islam and Arabic bachelor, I wanted to investigate what is happening in other countries with respect to Islamic banking. This thesis was the perfect opportunity to combine my existing knowledge about (Islamic) banking and my newly obtained analytical skills from the Master. During my thesis, I had the opportunity to do an internship at the Rabobank. I learned a lot during my internship and met a lot of nice and helpful people. I want to thank Barbara Baarsma for giving me the opportunity to do the internship and for her great feedback. Also, I want to thank Raphie Hayat for his feedback and help during the internship. Furthermore, I want to thank my family, friends and especially my boyfriend for all the support, help and good advice that I needed during this period.

I hope you will enjoy reading this thesis.

(4)

4

Chapter: 1 Introduction

Islamic banking is becoming bigger and is getting more attention, mostly during the last couple of years. Islamic banking is quite different from conventional banking, because come practices which are common in conventional banking, like charging interest, are prohibited in Islamic banking. In other words, Islamic banks are not allowed to charge interest on loans and those banks are also not allowed to offer a fixed rate of return on deposits (Chong & Liu, 2009).

Furthermore, trade or investing in, for example, alcohol, drugs and weapons is prohibited in Islamic banking. Thus, Islamic banking could be considered as a form of ethical banking (Tariq, 2004). However, not every product that is prohibited in Islamic banking is also prohibited in ethical banking, like for example pork. Because of these rules, Islamic banking caters to the Muslim demographic, which might make it more attractive for Muslims compared to standard products. This might help Islamic banks to penetrate these markets and increase competition in existing markets.

Currently there is no consensus on the profitability, liquidity and efficiency of Islamic banking. Derbel, Bouraoui, & Dammak (2011)believe that Islamic banking could have been the solution to the financial crisis in 2008, since it would be more stable than the conventional banking system. But Islamic banking may not be as efficient as conventional banking. Yudistira (2013) found out that Islamic banking is less efficient than conventional banking. However, it remains interesting to investigate whether Islamic banking affects competition in a country. Is there more competition with Islamic banks or is there a different reason why Islamic banking can earn a market share in a country?

In this thesis the competition between Islamic and conventional banks was investigated, with the main focus on the banking sector in Malaysia. Malaysia is reported to have the largest Islamic banking, capital and insurance market in the world (Furqani & Mulyany, 2009) and Malaysia is operating on basis of the dual banking system, which means that Islamic and conventional banks coexist (Chong & Liu, 2009). The fact that Malaysia works on the basis of this dual banking system and has the largest Islamic banking sector allows for easy comparison between Islamic and conventional banks.

This research contributes to the existing literature by using a richer, more recent dataset in comparison with other papers. In this thesis, the competition will be compared with conventional banks which might be interesting for other countries that are willing to introduce Islamic banking.

(7)

7 A hypothesis is that an increase in the number of players in the market, increases the competition. However, the Islamic banking sector is heavily subsidized and helped by the government (Trotsenburg, 2013), which may give the Islamic banks unfair advantages in comparison with conventional banks. The aim of this thesis is to investigate whether Islamic banks in Malaysia create more competition or not. Therefore, the research question of this thesis is:

Has Islamic banking increased the competition in the banking sector in Malaysia?

To be able to answer the research question, the following sub-questions will be answered: 1. What is Islamic banking and what are the principles?

2. How has Islamic banking developed in Malaysia?

3. What are the characteristics of the banking sector in Malaysia, in particular the Islamic banking sector?

4. How is competition between banks measured?

5. Are these measures equally useful when measuring competition between Islamic banks? And if not, what other measures are useful?

The remainder of this thesis is organized as follows. Chapter 2 provides an overview of the literature related to Islamic banking in Malaysia. Chapter 3 provides a description of the concepts and practices of Islamic banking. In chapter 4 the development of Islamic banking in Malaysia will be outlined. In chapter 5 the Malaysian banking sector will be explained based on the Structure-Conduct-Performance model. Chapter 6 will explain the methodology and results, which is followed by a discussion of the methodology and results in chapter 7. At last in chapter 8, there will be a conclusion and the research question will be answered.

(8)

8

Chapter 2: Literature review

A few studies have been conducted in order to analyze competition between banks in Malaysia.

Majid, Zulkhibri and Sufian (2007) examined the competition between Malaysian Islamic banks from 2001 to 2005. They used the H-statistic of the Panzar and Rosse model together with the ROAs of the Islamic banks. They conclude that competition between Islamic banks has increased in the period they analyzed, based on the market structure which was perfectly competitive, and the rejection of a monopoly in this banking sector. Their results also suggest that competitive behavior of banks not only depends on the number of banks in the sector but could also be the result of other factors, for example low entry barriers. If there is a treat of entry, the incumbents may act more competitive than they would when there are high entry barriers.

Weill (2011) compared the market power of Islamic and conventional banks from 17 countries where Islamic and conventional banks coexist. Data over the period from 2000-2007 was used for this research. It is based on the Lerner index and as a robustness check, the Panzar and Rosse model was used. There was no statistical difference in market power between both banking sectors, but by including control variables in the Lerner index, Islamic banks had significantly less market power than conventional banks. After conducting additional robustness checks, it was shown that Islamic banks do not differ in competitiveness from conventional banks.

Another research also based on the Lerner index and the Panzar and Rosse model in which data is used of 13 countries with Islamic and conventional banks, showed that Islamic banks use more of their assets in financing than conventional banks, and that these Islamic banks are more capitalized. However, the research also found that Islamic banking is less competitive than conventional banking (Ariss, 2010).

With respect to efficiency of Islamic banking in Malaysia, Mokhtar, Abdullah and

Al-Habshi (2006) used a stochastic frontier approach. They used data over the period 1997-2003 and their study tested the technical and cost efficiency of the Islamic banks. The study shows that the efficiency of Islamic banking has been increasing over the time, while the efficiency of commercial banks remained stable. However, the efficiency of Islamic banks was still below that of commercial banks. They studied also the difference in efficiency between Islamic windows and full-fledged Islamic banks and they found out that full-fledged Islamic banks are more efficient that Islamic windows.

(9)

9

Chapter 3: Islamic banking and its principles

In this chapter the first sub-question will be answered: what is Islamic banking and what are

its principles? This question will be answered by describing the concepts and practices of

Islamic banking, history, legislation and its enforcement will be explained. Also, the most common Islamic financial products will be discussed.

3.1 Islamic banking

In traditional banking there are some common practices when conducting business, like charging interest to cover the risk of a counterparty defaulting on a loan. However, this practice is prohibited by Islamic law, called the Sharia (Talal-Azimi, 2011). This has major implications for the products offered by Islamic banks, since many common practices (in the financial sector) are prohibited. Table 1 gives an overview of practices which are forbidden in Islamic finance1_.

Prohibited Definition

Riba al-fadl Trade in similar goods of different quality. For example, trade 1kg of high quality dates for 3kg low quality dates.

Riba al-nasia A fixed return on a loan (interest), meaning that the majority of the risk is carried by the borrower. This applies to payments made over time.

Gharar Speculation or gambling, where variance in outcome for one party can harm either party. For example, a merchant paying a fixed sum up front for a daily yield of a fisherman.

Maysir Excessive risk taking, like investing in derivatives or options. Haram products Goods that are not allowed for Muslims, like pork and alcohol.

Table 1 Forbidden practices in Islamic banking (Own summary)

Furthermore, one may not gain from transactions without putting in any work or effort, which is the case when a bank lends money and requires a fixed return on this loan.

• One may not shift the risk of one party to another, but risk must be shared among the parties involved (Chong and Liu, 2009), and second;

• financial transactions must have real economic activity underlying them (Chong and Liu, 2009).

(10)

10 3.2 Sharia boards

In order to enforce the Sharia, all institutions that provide Islamic products have a so-called

Sharia board. These boards have an advisory and supervisory role. The board will evaluate

whether or not the companies follow the rules of Islamic finance and their decree is binding. The boards always have a minimum of three Sharia scholars and consist out of an uneven number to be always able to have a majority of the votes (Talal-Azimi, 2011).

The disadvantage of decentralized Sharia boards per institution, is the fact that Sharia scholars differ in their school of thought, which may vary within a country or even between institutions. They all have their own distinguished point of view, but there is no consensus on what the best product is or not (Visser, 2009). Thus, some banks will offer products that are accepted by their own Sharia board but may not be accepted by other Sharia boards. Also, another disadvantage is the fact that Sharia scholars are scarce and that this market is very concentrated, which makes the Sharia supervision very expensive (Hayat, Den Butter and Kock, 2013).

To create a uniform standard for the financial products, there are a number of standard-setting bodies. In 1990 the Acounting and Auditing Organisation for Islamic Financial Institutions (AAOIFI) was established in Bahrein. The AAOIFI issues standards on how to follow the Islamic law to make sure that the products are the right ones, it is focused on accounting and auditing (Hayat, Den Butter and Kock, 2013). Their goal is to create clear standards and rules for Islamic finance products (AAOIFI, 2017).

Another setting body is the Malaysian Islamic Financial Services Board (IFSB). This board was established in 2002 in Malaysia. The IFSB has a similar function as the AAOIFI, but focusses more on corporate governance and risk management (Hayat, Den Butter and Kock, 2013). The mission of the IFSB is to guarantee that there is stability in Islamic finance products (IFSB, 2010).

So, the setting bodies offer clarity regarding the Islamic financial products in order to make it easier for the Sharia boards to evaluate whether the bank is following the right principles and is offering the appropriate financial products.

3.3 Common products in Islamic banking

Even though many practices are prohibited in Islamic finance, the majority of the financial products offered by conventional banks are also offered by Islamic institutions. However, these

(11)

11 products are adjusted in such a way that they fall within the boundaries of the Sharia but limit the downsides for customers.

Product Definition

Mortgages

Murabaha Mark-up financing: Bank buys a house from party A and sells this house to party B for the initial price plus a mark-up to cover the transaction costs. Party B can pay the home at once or in terms.

Musharaka mutanaqisa

Diminishing partnership: Bank and buyer of a house become co-owner of this house, until the buyer has paid off the entire loan.

Ijara wa-Iqtina Lease purchase: Bank buys a home and the borrower makes a monthly payment to make use of the real estate. If the borrower wants to buy the home at the end of the contract, she can do so at a predetermined price.

Loans

Bay-inah Sale-repurchase: Bank buys an asset of a client, so this client can obtain a loan. Afterwards, the client buys this asset back at a predetermined price, which includes a risk premium in the form of a mark-up.

Tawarruq Three party sale: Bank buys an asset from client A and sells it to client B with a mark-up. Afterwards, client B sells the asset back to client A without the markup.

Financial certificates, bonds

Sukuk Islamic bonds: Asset-backed medium-term notes. Client who wants to borrow using these bonds has to have full ownership of the underlying asset. Cannot be issued against money or debt.

Insurances

Takaful Insurance: Insurance fund with voluntary contributions, based on the desired coverage, like duration and risk coverage. The takaful operator does the administration and all remaining funds are used as insurance. Customers are not insured against tail risk.

Table 2 Common products in Islamic banking (Own summary)

3.4 Implications for Islamic financial institutions

Since Islamic banks only offer products which differ from conventional banks, this also has some implications for their balance sheet and the way they compete with other financial institutions. As a result, one might expect that Islamic banks differ significantly from

(12)

12 conventional banks with respect to profitability, liquidity and efficiency. However, the outcomes in the literature are mixed, so no conclusion can be drawn at this point.

Chapter 4: Development Islamic banking in Malaysia

In this chapter the second sub-question will be answered: How has Islamic banking developed

in Malaysia? First, the introduction of Islamic banking in Malaysia will be discussed.

Afterwards, there will be a review about the change of their market position. Finally, there will be explained how Islamic banking got a bigger market share.

4.1 History of Islamic banking in Malaysia, the first Islamic bank

Islamic banking was first introduced in Malaysia in 1963. In this year, the Malaysian government introduced the Pilgrims Management Fund and Board of Malaysia. The purpose of this fund was to facilitate Muslims who intend to go to Mecca to perform their hajj (pilgrimage). The pilgrimage to Mecca was made possible by pooling and investing government savings. This establishment is regarded as the first Sharia based banking product (Mokhtar, Abdullah and Al-Habshi, 2006), because it was based on Islamic principles and savings were used for the pilgrimage to Mecca.

Muslims in Malaysia were demanding Sharia compliant products. Therefore, the government of Malaysia started a committee in 1982 to investigate the possibility of having these banking products in Malaysia. In response to that and the enactment of the Islamic Banking Act in 1983, the first Islamic bank in Malaysia was founded on the first July of 1983, Bank Islam Malaysia Berhad (BIMB). BIMB was given a starting capital of RM580 million. These funds partially came from the government, the pilgrim funds and some religious counsels and agencies (Sukmana and Kassim, 2010).

BIMB performed well since its establishment. Four years later, in 1987, BIMB ranked twelfth among twenty-three local commercial banks in terms of assets and deposits. In terms of shareholder’s funds, BIMB ranked ninth of the twenty-three banks. The market share, which was in terms of deposits, increased from a half percent in 1984 to more than 2 percent in 1988 (Ariff, 1989).

4.2 After the first Islamic bank

BIMB was the only Islamic bank in Malaysia until 1993. In this year the Ministry of Finance introduced the Interest-free Banking Scheme for conventional banks. At first, this scheme was only for three banks in Malaysia; Malayan Banking Berhad, Bank Bumiputera and United

(13)

13 Malayan Banking Cooperation to expand Islamic banking and finance products (Mohamad, Abdullah, Mohamad and Abidin, 2013).

This was a turning point since this scheme made it possible for conventional banks to offer Islamic banking products. However, this is only allowed when Islamic products are being offered completely separately from the conventional products, which means separate balance sheets (Mohamad, Abdullah, Mohamad and Abidin, 2013). As a result of this scheme, Islamic banking and finance grew substantially in Malaysia (Mokhtar, Abdullah and Al-Habshi, 2006). Malaysia was the first country that practices this dual banking model, which means that Islamic and conventional banks coexist. Some countries choose to only bank the Islamic way, while other countries only bank in the conventional way, and others have the “conventional plus” system. The last is a system where there are conventional banks with a few Islamic institutions, for example as an Islamic window (TKBB, n.d.).

In order to maintain this position as a country that practices the dual banking system, the national bank of Malaysia, Bank Negara Malaysia (BNM), introduced in 1994 Islamic Interbank Money Market (IIMM). Their goal was to create an Islamic banking system that could maintain its position side-by-side the conventional banking system (TKBB, n.d.).

4.3After the introduction of IIMM

In 1999, the government approved the founding of another Malaysian Islamic bank. This second Islamic bank was a merger of two banks in Malaysia, namely Bank Bumiputera Malaysia Berhad (BBMB) and Bank of Commerce Malaysia Berhad (BOCB). This merger took place when both of these banks started with an Islamic banking section and made one institute of it (Mohamad, Abdullah, Mohamad and Abidin, 2013). With the entry of Bank Mumalat Malaysia Berhad (BMMB) (TKBB, n.d.), the BIMB lost their monopoly.

After the establishment of BMMB, BNM issued three additional Islamic banking licenses in 2004. These licenses were for the international banks Al Rajhi Banking & Investment Corporation, a consortium led by Qatar Islamic Bank and Kuwait Finance House (Bank Negara Malaysia, 2004). Since the Malaysian government started to provide more banking licenses, the market was opened up to new entrants. Since it became possible to obtain a license, all the banks in Malaysia that opened their own Islamic window, were encouraged to also open Islamic subsidiary’s that could eventually be full-fledged Islamic banks (Mohamad, Abdullah, Mohamad and Abidin, 2013).

With respect to the sukuk (bonds) market in Malaysia, the International Islamic Liquidity Management Corporation (IILM) was established in 2010. The IILM was established

(14)

14 in order to make it possible to create and issue short-term Sharia compliant financial products (Cheng & Lim, 2017).

(15)

15

Chapter 5 Malaysian banking sector

In this chapter the third sub-question will be answered: What are the characteristics of the

banking sector in Malaysia, in particular the Islamic banking sector? The Malaysian banking

sector will be described by using the Structure-Conduct-Performance (SCP) model. The structure will illustrate the supply and the demand side of the market and entry and exit barriers. The performance regards the growth, profitability and efficiency of the market.

5.1 Structure of Malaysian banking sector

Market characteristics determine the structure of the market. Factors such as the number of firms, the size of those firms, the barriers to entry, the demand for the products that are being offered are important determinants for market competition.

5.1.1 Overview Malaysian banks, supply structure

First, all the banks of Malaysia are depicted in the following tables. This information was obtained from the Malaysian central bank (Bank Negara Malaysia, n.d. b). Table 3 is an overview of the dual institutions and table 4 concerns the Islamic banks and the conventional banks. The banks are listed alphabetically and the table does not address individual bank characteristics, for example growth or assets. Furthermore, bank ownership is denoted by L for local owned banks and F for foreign owned banks.

Table 3 Malaysian dual institutions (Bank Negara Malaysia, n.d. b)

Islamic banks Malaysia Ownership Conventional banks Malaysia Ownership

Affin Islamic Bank Berhad L Affin Bank Berhad L

Alliance Islamic Bank Berhad L Alliance Bank Malaysia Berhad L

AmBank Islamic Berhad L AmBank (M) Berhad L

CIMB Islamic Bank Berhad L CIMB Bank Berhad L

HSBC Amanah Malaysia Berhad F HSBC Bank Malaysia Berhad F

Hong Leong Islamic Bank Berhad L Hong Leong Bank Berhad L

Maybank Islamic Berhad L Malayan Banking Berhad L

OCBC Al-Amin Bank Berhad F OCBC Bank (Malaysia) Berhad F

Public Islamic Bank Berhad L Public Bank Berhad L

RHB Islamic Bank Berhad L RHB Bank Berhad L

(16)

16 Table 4 Malaysian banks (Bank Negara Malaysia, n.d. b)

Figure 1 Top 10 Malaysian banks, based on assets (Bureau van Dijk, 2017)

Figure 1 contains the top 10 of Malaysian banks based on the total assets on the balance sheet. This figure shows that the majority of the large banks in Malaysia are conventional banks. There are only two Islamic banks in the top 10, namely Maybank Islamic Berhad with 49.85 billion USD in assets and CIMB Islamic Berhad with 20.99 billion USD in assets (Bureau van Dijk, 2017).

Islamic banks Malaysia Ownership Conventional banks Malaysia Ownership

Al Rajhi Banking & Investment Corporation (Malaysia) Berhad F BNP Paribas Malaysia Berhad F

Bank Islam Malaysia Berhad L Bangkok Bank Berhad F

Bank Muamalat Malaysia Berhad L Bank of America Malysia Berhad F Kuwait Finance House (Malaysia) Berhad F Bank of China (Malaysia) Berhad F MBSB Bank Berhad L China Construction Bank (Malaysia) Berhad F

Citibank Berhad F

Deutsche Bank (Malaysia) Berhad F India International Bank (Malaysia) Berhad F Industrial and Commercial Bank of China (Malaysia) BerhadF J.P. Morgan Chase Bank Berhad F MUFG Bank (Malaysia) Berhad F Mizuho Bank (Malaysia) Berhad F National Bank of Abu Dhabi Malaysia Berhad F Sumitomo Mitsui Banking Corporation Malaysia BerhadF The Bank of Nova Scotia Berhad F United Overseas Bank (Malaysia) Bhd. F

125.471.888 77.219.291 75.936.581 49.851.071 43.778.926 38.383.020 22.384.235 20.993.958 _20.406.149 18.275.033 Malayan Banking Berhad Public Bank Berhad CIMB Bank Berhad Maybank Islamic Berhad RHB Bank Berhad Hong Leong Bank Berhad United Overseas Bank (Malaysia) bhd. CIMB Islamic Berhad AmBank (M) Berhad OCBC Bank (Malaysia) Berhad

(17)

17

Figure 2 Assets 2017 of dual institutions (Bureau van Dijk, 2017)

The total assets of the dual institutions are depicted in figure 2. From this figure it becomes clear that there is a large disparity between total assets of the Islamic branch and the conventional branch of the same holding. On average, just 20 percent of total assets is held by the Islamic branch of the bank.

5.1.2 Muslim population, demand structure

In order to determine the demand for (Islamic) financial products, a clear overview of the Malaysian population is required. Therefore, we have to look at the percentage of the Muslim population in Malaysia and, more importantly, the percentage of that population interested in Islamic banking and finance products and those who are indifferent between Islamic and standard banking products. Protentional customers for Islamic banking products could also be Muslims. If financial products are homogenous, but the Islamic version is cheaper, non-Muslims might switch from conventional banking products to Islamic banking products.

According to a study done by Pew Research Center (2010), the Muslim population in Malaysia is expected to grow substantially faster than any other religion in Malaysia, namely from 28.400.000 Muslims in 2010 to 45.190.000 in 2050. This is an increase of 8,7 percentage point (from 63,7 percent Muslim to 72,4 percent).

With the expected growth in Muslim population, one would expect increasing demand for Islamic banking and finance products. The Malaysian Reserves states that:

“unfortunately, no one has spent any serious amount of their research and development budget

to study the demand for Islamic banking in the country. Most of the Islamic banking strategies

0 20.000.000 40.000.000 60.000.000 80.000.000 100.000.000 120.000.000 140.000.000 Affin Bank Allia nce B ank AmBa nk CIM B Ba nk HSBC Bank Hong Leong Bank Mala yan/ Mayba nk OCBC Bank Publ ic B ank RHB bank Standa rd Cha rtered

Dual institutions assets (x 1000 USD)

Islamic assets Conventional assets

(18)

18

are at best based on some sophisticated in-house guesswork or some incomplete, incorrect and potentially misleading research by the leading consultancy firms.”

Thus, it is impossible to draw any conclusions about the demand for Islamic banking and finance products because it is based on rough guesses and incomplete data.

This means that the growing size of Muslim population in Malaysian is, at best, a very rough proxy for the total demand for Islamic banking products. Furthermore, demand also depends on public awareness about Islamic banking products. This means that more data should be gathered regarding awareness and actual demand on the household level. This can be done by using a nationwide survey. In this survey researchers could determine the willingness to pay of agents with respect to Islamic banking products. In other words, customers might be willing to switch to Islamic finance products, even when the price is higher based on their willingness to pay. This information equips bankers with the tools to determine total demand.

5.1.3 Regulation Malaysian banking sector

BNM published in 2011 the Financial Sector Blueprint (FSB). This was intended to determine the direction of the Malaysian financial sector over a period of 10 years in order for Malaysia to become a high value-added, high income economy (The Law Review, 2016). In line with the FSB, the Financial Services Act 2013 (FSA) and the Islamic Financial Services Act 2013 (IFSA) were introduced. These two acts were introduced in order to have a regulatory framework for the conventional and the Islamic banking and finance sectors. Also, it gave BNM greater legal power to limit future risks to financial stability, to promote competition and to increase consumer protection. The acts also provide guidelines, directions and legislation in order to obtain an Islamic banking license (The Law Review, 2016).

BNM has the legal power to operate as regulator and supervisor of banks licensed under the FSA and IFSA. BNM wants to promote monetary and financial stability in order to help the Malaysian economy grow at a sustainable rate. They want to achieve this by focusing their monetary policy on price stability. BNM also advises the government on public debt and macroeconomic policies. The central bank is also the sole authority in issuing money and managing the international reserves of Malaysia (Bank Negara Malaysia, n.d. a).

The Malaysia Competition Commission (MyCC) is founded to enforce the Competition Act 2010 (CA 2010) with the main role to facilitate and protect competitive behavior for the benefit of the firms, consumers and the economy as a whole (MyCC, n.d.). The CA 2010 is similar to article 101 of the Treaty on the Functioning of the European Union (TFEU). The CA 2010 prohibits certain practices in order to increase competition. Companies may not abuse

(19)

19 their market position and anti-competitive behavior, like collusion, is prohibited. It is a relatively new act in Malaysia, since it came in effect in 2012 (See, 2015), so nothing can be said about the effectiveness of this law just yet. However, some initial positive results have been found regarding the prevention of price fixing. But there are still many challenges, especially when it comes to more complex cases. It might even be necessary to expand the current legislation with additional rules, for example regarding mergers (Lee, 2017). The first positive signs regarding the CA 2010 are there, but conclusions about the effectives cannot be drawn just yet. However, the effect of the CA 2010 legislation is expected to foster competition.

5.1.4 Entry and exit barriers

The entry of new banks into the Malaysian market is regulated by the FSA and the IFSA. Therefore, this market is strongly regulated by the BNM. When a firm wants to enter the market, it has to be a public company incorporated in the Companies Act 2016 (CA 2016). In order to get a license, the possible entrant has to send in a written application to the Minister by submitting this through BNM. Afterwards, BNM will look into the application based on the type of institution, FSA for conventional banks and the IFSA for Islamic bank. The BNM determines whether or not all criteria are met and if this new entrant will contribute to competition in the banking sector. Based on the outcome of this evaluation, the BNM will give advice to the Minister (Luk, Foong and Heung, 2017). These strict rules function as entry barriers, limiting the number of players in the market. In the end, this could negatively affect the degree of competition in a market.

5.2 Conduct

The market conduct of Malaysian banks can be analyzed by using marketing expenditures. The total marketing budget can be used as a proxy for market penetration in a given year. However, since there is no data available on the marketing expenditures on the individual level, the market conduct of Malaysian banks cannot be analyzed.

(20)

20 5.3 Performance

In table 5 some characteristics of the Islamic banking sector in Malaysia are depicted. Malaysia Global Percentage Islamic Finance assets (US$ Mn) 415.418 1.814.086 23% Islamic banking assets (US$ Mn) 173.956 1.345.891 13%

Takaful/Retakaful assets (US$ Mn) 8.205 33.390 25%

Number of Takaful/Retakaful Operators 21 308 7%

Value of Outstanding Sukuk (US$ Mn) 167.256 295.094 57% Net asset value of Islamic funds (US$ Mn) 17.797 55.794 32%

Table 5 Characteristics Islamic banking sector (Thomson Reuters, 2015)

It is clear from table 5 that a large share of all the Islamic finance products sold globally are being sold in Malaysia. In some cases, Malaysia accounts for more than half of the products outstanding globally, for example with sukuk. As a result, Malaysia has the largest Islamic finance economy of the world (Thomson Reuters, 2015).

5.3.1 Growth of banks

In order to determine the growth of banks, asset growth was used as a proxy. Growth was measured over 7 years (2011 – 2017). Figure 3 depicts the total assets of the 10 largest banks in Malaysia.

Figure 3 Growth assets of top 10 Malaysian banks (Bureau van Dijk, 2017)

0 20.000.000 40.000.000 60.000.000 80.000.000 100.000.000 120.000.000 140.000.000 2010 2011 2012 2013 2014 2015 2016 2017 2018 T ot al as se ts Year

Asset growth, top 10 banks Malaysia (x 1000

USD

)

Malayan Banking Berhad Public Bank Berhad CIMB Bank Berhad Maybank Islamic Berhad RHB Bank Berhad Hong Leong Bank Berhad United Overseas Bank (Malaysia) bhd. CIMB Islamic Berhad AmBank (M) Berhad OCBC Bank (Malaysia) Berhad

(21)

21

Figure 4 Asset growth Islamic banks (Bureau van Dijk, 2017)

Figure 5 Asset growth Islamic banks, excl. MIB and CIMB (Bureau van Dijk, 2017)

Figure 4 shows the asset growth of all the Islamic banks in Malaysia. One bank, the Maybank Islamic Berhad (MIB), has substantially more assets than other Islamic banks and is one of the two Islamic banks in the top 10 largest banks in Malaysia. The two largest Islamic banks have been omitted in Figure 5 to give a clearer view of the growth of assets among the other banks. The total assets held by the top 10 Malaysian banks and all the Islamic banks declined in the year 2014/2015 but increased again 2016/2017. The reason for this decline is likely the sharp drop in oil prices mid-2014. At this point the price of oil was at the lowest level since 2009 (Chandran, 2015). The lower oil prices have negatively affected the economies of the

0 10.000.000 20.000.000 30.000.000 40.000.000 50.000.000 60.000.000 2010 2011 2012 2013 2014 2015 2016 2017 2018 Assets Year

Asset growth Islamic banks ( x1000 USD) Affin Islamic Bank Berhad

Alliance Islamic Bank Berhad AmBank Islamic Berhad CIMB Islamic Bank Berhad HSBC Amanah Malays ia Berhad Hong Leong Islamic Bank Berhad Maybank Islamic Berhad OCBC Al-Amin Bank Berhad Public Islamic Bank Berhad RHB Islamic Bank Berhad Standard Chartered Saadiq Berhad Al Rajhi Banking & Investment Corporation (Malaysia) Berhad

Bank Islam Malaysia Berhad Bank Muamalat Malaysia Berhad Kuwait Finance House (Malays ia) Berhad

0 2.000.000 4.000.000 6.000.000 8.000.000 10.000.000 12.000.000 14.000.000 16.000.000 2010 2011 2012 2013 2014 2015 2016 2017 2018 Asse ts Year

Asset growth Islamic banks (x 1000 USD) (excl. MIB and CIMB)

Affin Islamic Bank Berhad Alliance Islamic Bank Berhad AmBank Islamic Berhad HSBC Amanah Malays ia Berhad Hong Leong Islamic Bank Berhad OCBC Al-Amin Bank Berhad Public Islamic Bank Berhad RHB Islamic Bank Berhad Standard Chartered Saadiq Berhad Al Rajhi Banking & Investment Corporation (Malaysia) Berhad

Bank Islam Malaysia Berhad Bank Muamalat Malaysia Berhad Kuwait Finance House (Malays ia) Berhad

(22)

22 middle eastern countries since a large share of their income comes from refining oil. This might explain why the Islamic banks have suffered, since a large share of their customer base is in this region (Cunado, Jo, and de Gracia, 2015).

5.3.2 Profitability and efficiency of banks

For the profitability and efficiency of banks the Return on Assets (ROA) and the Return on Equity (ROE) can be used.

𝑅𝑂𝐴_$% = '(% *+,-.(/0

1-%23 455(%5/0 (1) ROA is calculated for all the banks by the ratio of the net income of the last reported year by the total value of the assets in the same year. The ROAs of the banks are being compared with each other. In general, ROAs over 5 percent are being considered as good ROAs. Furthermore, when the ROA is higher, the company is able to use their resources well in order to generate profit/income. This implies that those banks use their assets more efficiently (Kennon, 2018).

The second measure, ROE, shows how well a company can generate returns based on initial funds obtained from investors. In other words, investors prefer a high ROE, since it implies that a company is using their investments efficiently (Kennon, 2017).

𝑅𝑂𝐸_$% = '(% *+,-.(/0

1-%23 7829(8-3:(95 ;<=$%>/0 (2) ROE is calculated for all the banks by the ratio of the net income of the last reported year by the total value of the shareholders’ equity in the same year (Kennon, 2017).

Table 6 ROAs and ROEs Malaysian dual banks (Bureau van Dijk, 2017)

Islamic banks Malaysia ROA ROE Conventional banks Malaysia ROA ROE

Affin Islamic Bank Berhad 0,74% 9,73% Affin Bank Berhad 0,58% 3,77%

Alliance Islamic Bank Berhad 0,77% 9,01% Alliance Bank Malaysia Berhad 1,03% 10,55%

AmBank Islamic Berhad 0,70% 8,81% AmBank (M) Berhad 1,04% 11,23%

CIMB Islamic Bank Berhad 0,75% 13,99% CIMB Bank Berhad 1,18% 11,48%

HSBC Amanah Malaysia Berhad 0,49% 5,34% HSBC Bank Malaysia Berhad 1,27% 10,21% Hong Leong Islamic Bank Berhad 0,78% 11,01% Hong Leong Bank Berhad 1,06% 9,46%

Maybank Islamic Berhad 0,86% 18,66% Malayan Banking Berhad 1,20% 9,84%

OCBC Al-Amin Bank Berhad 1,21% 14,35% OCBC Bank (Malaysia) Berhad 0,80% 11,23%

Public Islamic Bank Berhad 0,72% 9,80% Public Bank Berhad 1,61% 15,50%

RHB Islamic Bank Berhad 0,67% 11,09% RHB Bank Berhad 1,10% 10,07%

(23)

23

Table 7 ROAs and ROEs Malaysian Islamic and conventional banks (Bureau van Dijk, 2017)

Table 6 and table 7 depict the ROAs and ROEs of all the banks in Malaysia in 2017. The ROA of all Malaysian banks was quite low. Almost all the Islamic banks, except OCBC Al-Amin Bank Berhad, have a ROA below 1 percent. In the case of the conventional banks, it tends to vary more. To be able to compare the different ROAs, the average of both banking systems is calculated for each year and across years. The average ROA of the Islamic banks is 0,65 percent (0,63 percent on average from 2011 – 2017) and of the conventional banks 0,87 (0,85 percent on average from 2011 – 2017) percent.

In order to test the differences between the ROAs and ROEs, a Kolmogorov-Smirnov test was conducted (Smirnov, 1939), which compares the distributions of the ROAs and ROEs between both groups (Islamic banks vs. conventional banks). This test does not require the data to be normally distributed since it is a non-parametric test. Furthermore, one of the assumptions underlying this method is that the two groups are independent. This assumption holds since the balance sheets have to be completely separate in order to obtain an Islamic banking license for an Islamic window.

Table 8 Outcomes Kolmogorov-Smirnov tests ROA

(1) (2) (3) (4) (5) (6) (7) 2011 2012 2013 2014 2015 2016 2017 Conventional 0.000 0.000 0.1333 0.1710 0.0533 0.0400 0.000 (1.000) (1.000) (0.737) (0.588) (0.948) (0.970) (1.000) Islamic -0.8000*** -0.6083** -0.4333** -0.3913* -0.4400** -0.4267** -0.5833** (0.002) (0.021) (0.040) (0.062) (0.027) (0.033) (0.017) Combined KS 0.8000*** 0.6083** 0.4333* 0.3913 0.4400* 0.4267* 0.5833** (0.004) (0.042) (0.080) (0.124) (0.053) (0.066) (0.034) P-value in parentheses *** p<0.01, ** p<0.05, * p<0.1

Islamic banks Malaysia ROA ROE Conventional banks Malaysia ROA ROE

Al Rajhi Banking & Investment Corporation (Malaysia) Berhad 0,12% 1,45% BNP Paribas Malaysia Berhad 0,31% 1,88%

Bank Islam Malaysia Berhad 0,95% 12,11% Bangkok Bank Berhad 0,74% 3,84%

Bank Muamalat Malaysia Berhad 0,64% 7,01% Bank of America Malysia Berhad 0,99% 5,05% Kuwait Finance House (Malaysia) Berhad 0,07% 0,41% Bank of China (Malaysia) Berhad 1,23% 8,82%

MBSB Bank Berhad - - China Construction Bank (Malaysia) Berhad -

-Citibank Berhad 1,41% 13,26%

Deutsche Bank (Malaysia) Berhad 1,24% 8,38% India International Bank (Malaysia) Berhad -0,29% -0,43% Industrial and Commercial Bank of China (Malaysia) Berhad0,83% 3,54% J.P. Morgan Chase Bank Berhad 0,89% 7,96% MUFG Bank (Malaysia) Berhad -

-Mizuho Bank (Malaysia) Berhad 0,21% 2,30% National Bank of Abu Dhabi Malaysia Berhad 0,09% 0,19% Sumitomo Mitsui Banking Corporation Malaysia Berhad 0,45% 4,37% The Bank of Nova Scotia Berhad 1,05% 2,66% United Overseas Bank (Malaysia) Bhd. 1,10% 13,11%

(24)

24

Table 9 Outcomes Kolmogorov-Smirnov tests ROE

(1) (2) (3) (4) (5) (6) (7) 2011 2012 2013 2014 2015 2016 2017 Conventional 0.000 0.0714 0.3333 0.2857 0.2912 0.2418 0.1667 (1.000) (0.949) (0.155) (0.236) (0.214) (0.345) (0.717) Islamic -0.6429** -0.3571 -0.0714 -0.1310 -0.0934 -0.0714 -0.1667 (0.021) (0.273) (0.918) (0.738) (0.853) (0.911) (0.717) Combined KS 0.6429** 0.3571 0.3333 0.2857 0.2912 0.2418 0.1667 (0.042) (0.535) (0.308) (0.466) (0.423) (0.662) (0.996) P-value in parentheses *** p<0.01, ** p<0.05, * p<0.1

Looking at the outcomes depicted in table 8, the ROA is significantly higher for conventional banks compared to Islamic banks. Only in the year 2014 the results were insignificant; however, this might be explained because of the sudden drop in assets possibly caused by a decline in the oil-price.

From this it might be concluded that the Islamic banking system is less able to use their assets to generate profit. And the opposite holds for conventional banks, which have a significantly higher ROA, which implies that they are better able to generate profit with the available assets.

The average ROE of the Islamic banks is 9,10 percent (9,35 percent on average from 2011 – 2017) and of the conventional banks 7,40 percent (8,26 percent on average from 2011 – 2017). Based on these percentages, one might assume that the Islamic banks are better able to use the investors’ funds to generate profits. However, from table 9 is clear that all tests but one, are insignificant. In this year, the ROE of conventional banks was significantly higher than the Islamic banks, which seems counterintuitive. But, since there were many missing observations in 2011 for conventional banks and the banks that reported their ROE had quite high ROEs, these results should be considered as biased.

(25)

25

Chapter 6: Measuring competition

In this chapter the fourth sub-question will be answered: How is competition between banks

measured? Several methods have been used to measure competition. These measures were

chosen since they can be applied to Islamic banks and allow for comparison between conventional banks and Islamic banks. Furthermore, using different measures can serve as a robustness check.

6.1 Herfindahl-Hirschman Index (HHI)

The HHI is a measure to calculate market concentration, which is calculated by summing the squared market shares of all the companies in the market.

𝐻𝐻𝐼_% = ∑+ (𝑀𝑆_$%)

$FG (3)

with HHI at time t, MS being the market share of the banks in the sector (Islamic/conventional/Malaysian) i at time t, and n being the number of the banks (Carbó, Humphrey, Maudos and Molyneux, 2009). The HHI varies between 0 to 10,000. With an HHI of 0, there is perfect competition and an HHI of 10,000 implies a monopoly. (The US Department of Justice, 2015). Between those values it varies not concentrated and highly concentrated. The market shares of the different sectors are based on the total assets of those sectors, thus the total Malaysian banking assets, the total Islamic banking assets and the total conventional banking assets. Balance sheet size is not exactly the same as the market share, but it is a commonly used as a proxy for the market share.

Market definition according to HHI value

General interpretation (Diallo and Tomek, 2015)

Horizontal Merger Guidelines (The US Department of Justice, 2015)

Highly competitive HHI ≤ 100 HHI ≤ 100

Not concentrated 100 < HHI < 1000 100 < HHI < 1500 Moderately concentrated 1000 < HHI < 1800 1500 < HHI < 2500

Highly concentrated HHI > 1800 HHI > 2500

Table 10 HHI value explanation

The smaller the value of the HHI, the less concentrated the market which means that there are many firms in the market and this may point towards more competition. Since the value of the HHI is quite high the market is highly concentrated, which means that there are less firms in the market, the market shares are concentrated around a few firms or both. In table 11 and figure

(26)

26 6 below the HHI values of the whole banking sector, the Islamic banks and the Commercial banks are depicted.

The HHI measure is used to as an instrument to evaluate possible mergers, since it focusses on market concertation. HHI values could be inaccurate because they do not take the competitive behavior of each bank into account. When banks are state owned they could be behaving differently than private or foreign banks. For example, large concentrations of state owned banks are considered less competitive. So, in this case, banks in the same sector could not be competing with each other (Berger, Demirguc-Kunt, Levine and Haubrich, 2004).

The HHI values in table 11 are calculated by using the share of total assets owned by the bank, which functions as a proxy for market share. This value is multiplied by 100 for each bank per year. Afterwards, this value is squared and summing all those values give the HHI per year. For the Malaysian banking sector all the banks are used, for the Islamic banking sector only the Islamic banks and for the conventional banking sector only the conventional banks. Due to the fact that there are more conventional than Islamic banks in Malaysia, the market concentration of the 10 largest banks (based on total assets) was also calculated. This is used as a robustness check, since the number of players in a market reduces market concentration.

Table 11 HHI values (Bureau van Dijk, 2017)

Figure 6 HHI values over time

HHI 2011 2012 2013 2014 2015 2016 2017

Malaysian Banking Sector 1113,53 1154,89 911,67 858,68 838,92 824,61 1168,93

Islamic Banks 1165,41 1219,31 1351,78 1342,97 1343,43 1450,44 2230,85

Conventional Banks 1007,84 1853,30 1372,52 1252,24 1243,92 1238,33 1750,71

Top 10 Islamic Banks 1139,26 1198,16 1332,36 1319,78 1321,77 1432,15 2208,26

Top 10 Conventional Banks 1007,61 1853,16 1337,11 1226,60 1220,14 1215,90 1734,59

0,00 500,00 1000,00 1500,00 2000,00 2500,00 2011 2012 2013 2014 2015 2016 2017 HHI Value

HHI Malaysia

Malays ian Banking Sector Islamic Banks Conventional Banks Top 10 Islamic Banks Top 10 Conventional Banks

(27)

27 For the whole Malaysian banking sector, the market is not concentrated based on the values of the HHI in 2015 and 2016. Those two years contain the information of all the banks in Malaysia. The values in 2011 and 2012 are much higher, due to the fact that there are a lot of missing observations. The same holds for 2017. In 2017. In other words, since the number of banks artificially goes down due to a lack of data, market concentration might be skewed based on fewer observations.

This can also be seen for only the conventional banks. In the years 2015 and 2016 we can see that the market is moderately concentrated. In 2014 the market is also moderately concentrated, but there is no information available for two banks in this sector. The banks (Bangkok Bank Berhad and Bank of China (Malaysia) Berhad) are relatively small with respects to their assets, so it did not change much in the value of the HHI. The peak in 2012 is due to the fact that there was no information available in 2011 for Malayan Banking Berhad, which is the biggest bank in Malaysia, significantly reducing market concentration as a result of this omission.

The last banking sector is the Islamic one. For the Islamic sector, all the data for the banks is available, except for some banks in 2017. This might explain the sharp increase in 2017 in comparison with 2016. However, the Islamic banking sector is moderately concentrated throughout the years and this remains stable around an HHI value of 1300.

6.2 Boone indicator

To measure the degree of competition, the Relative Profit Difference measure can be used, which will be referred to as the Boone indicator (Boone, 2008). The Boone indicator is calculated by regressing the log of profits on the log of the marginal costs. As a result, we get the elasticity between these two variables. The Boone model is characterized as follows:

ln(𝜋$%) = 𝛼$%+ 𝛽$%ln(𝑀𝐶$%), (4)

with π being the profits of the sector (Islamic/conventional/Malaysian) i at time t, and MC the marginal costs of the sector i at time t. The estimator β is the value for the Boone indicator. Following Schaeck and Cihák (2010) and Van Leuvensteijn, Bikker, Van Rixtel and Sørensen (2011), the profits will be calculated by using the ROA. We use the ROA as a proxy for profits because profits are difficult to measure. The ROA is calculated by dividing the net income of the bank by the total value of the assets. For the marginal costs, the average total costs (ATC) will be used because marginal costs cannot be observed directly. So, the formula will be:

(28)

28 ln(𝑅𝑂𝐴_$%) = 𝛼_$%+ 𝛽_$%ln(𝐴𝑇𝐶_$%). (5) For the ATC, the average costs of the bank as a share of total income are being used. Average costs include interest expenses and other operating expenses and the income consists of interest revenue and other operating income (overhead costs + loan loss provisions) (Schaeck and Cihák, 2010).

The reasoning behind the Boone indicator is that more efficient firms, which have lower costs, achieve higher profits. If the marginal costs increase, the profits of a company will decrease, so the expected sign of β will be negative. Thus, the more negative the indicator is, the higher the degree of competition. In other words, the higher the Boone indicator in absolute value, the higher the competition. A decrease, therefore, in absolute value of the Boone indicator means a decrease in the competition (World Bank, 2017).

Looking at the regressions in table 12, 13 and 14 below, it can be seen that the Boone indicator is statistically significant in all the cases. Almost all the results are significant on a 1 percent confidence level, only the value in 2011 for the Islamic and commercial banks is significant at the 10% level and the value in 2014 for the commercial banks is significant at the 5% confidence level.

The ROE in chapter 3 showed us that based only on the net income and the shareholders equity, Malaysian Islamic banks were expected to be more efficient than the conventional banks, but these results were insignificant. However, the values for the Boone indicator of the Islamic banks are higher in absolute value than the values for the conventional banks. Thus, one could say that based on the ROE and the Boone indicator, the Islamic banks might be more competitive than the conventional banks. This is also the case when the Boone indicator values of the Islamic banks are being compared to those of the Malaysian banking sector as a whole.

However, this conclusion has to be taken with caution. This conclusion only holds when the bank actually has higher profits as a result of lower costs. Otherwise, the higher profitability could be a sign of higher market power. This may come from banks that are being subsidized by the government with state-aid. Unfortunately, it was very difficult to find out which banks are getting state-aid and which are not, so a conclusion cannot be drawn based on this.

(29)

29 Table 12 Boone indicator Malaysian banks

(1) (2) (3) (4) (5) (6) (7) VARIABLES 2011 2012 2013 2014 2015 2016 2017 Boone -0.827*** -0.913*** -1.599*** -1.349*** -1.742*** -1.839*** -2.114*** (0.221) (0.187) (0.341) (0.325) (0.387) (0.293) (0.319) Constant -5.353*** -5.398*** -5.825*** -5.828*** -6.059*** -6.062*** -6.298*** (0.173) (0.157) (0.241) (0.227) (0.246) (0.186) (0.211) Observations 20 22 31 35 36 36 24 R-squared 0.438 0.544 0.431 0.343 0.374 0.536 0.667

Standard errors in parentheses *** p<0.01, ** p<0.05, * p<0.1

Table 13 Boone indicator Conventional banks

(1) (2) (3) (4) (5) (6) (7) VARIABLES 2011 2012 2013 2014 2015 2016 2017 Boone -0.584* -0.429*** -0.949*** -1.032** -1.659*** -1.656*** -1.594*** (0.242) (0.0965) (0.253) (0.400) (0.528) (0.371) (0.428) Constant -4.930*** -4.815*** -5.250*** -5.543*** -5.926*** -5.902*** -5.830*** (0.226) (0.103) (0.188) (0.294) (0.340) (0.248) (0.307) Observations 7 8 17 21 23 23 12 R-squared 0.539 0.767 0.483 0.260 0.320 0.487 0.582

Table 14 Boone indicator Islamic banks

(1) (2) (3) (4) (5) (6) (7) 2011 2012 2013 2014 2015 2016 2017 Boone -0.612* -1.301*** -2.295*** -2.163*** -2.087*** -2.162*** -2.312*** (0.288) (0.339) (0.655) (0.544) (0.400) (0.508) (0.474) Constant -5.332*** -5.702*** -6.403*** -6.425*** -6.412*** -6.320*** -6.523*** (0.200) (0.231) (0.433) (0.348) (0.250) (0.292) (0.284) Observations 13 14 14 14 13 13 12 R-squared 0.291 0.551 0.506 0.568 0.712 0.622 0.705

(30)

30 The Boone indicator in 2017 has a very high absolute value, even the highest of all the years, for the Islamic banks and the Malaysian banking sector as a whole. It could indicate that there was more competition in 2017, which implies that competition increased. The conclusion that could be drawn from the HHI is that the market in 2017 was more concentrated. The reason for this high concentration is because of a lack of data for multiple banks in 2017. This could bias the results and overestimate market concentration, which might not be the case if all data was available. The values of both measures contradict each other, but the reason for this could be because the HHI only looks at market shares and does not take other variables into account. It could be that there are less firms in the market, but it could still be a competitive market with less firms.

The conventional banks have a high Boone value in 2015 and 2016. These years are with respect to the HHI value the lowest in this sector (when the HHI of 2011 is not taken into account). This supports the hypothesis that sectors that are less concentrated are considered more competitive. So, the conventional market was the most competitive in these years based on the HHI value and the Boone indicator.

(31)

31

Chapter 7: Discussion

In this chapter the fifth sub-question will be answered: Are these measures equally useful when

measuring competition between Islamic banks? And if not, what other measures may be useful?

First, the limitations of the HHI and the Boone indicator will be discussed. Afterwards, some other measures will be explained that might be useful for measuring competition in Islamic banking.

7.1 Limitations of measures 7.1.1 HHI

Like stated earlier, the HHI can be inaccurate because it does not take competitive behavior of each bank into account. This is because it only measures market concentration, which does not give any evidence on the degree of competition in a market.

HHI is failing in explicitly defining the market. If there are 10 banks in one market that all have the same market share, say a market share of 10%. Based on the simple calculation of the HHI, the market would be not concentrated and therefore highly competitive. But it could be possible that one of those banks in the market is the only, or one of the only, bank who offers a certain product (Leon, 2015). This means that this bank has a high market share with respect to this product, which means that the bank has a monopoly with respect to that product. This is a limitation of the HHI since it only takes concentration at the macro level into account. It could be the case that this applies to Islamic finance, since they are the only institutions that offer Islamic products. Furthermore, not all Islamic banks offer all the Islamic banking products that are available, which makes the HHI even more unreliable. HHI may therefore not be a great tool for measuring competition between Islamic and non-Islamic banks. This could be interesting for further research, since it leaves room for improvement based on this specific measure. However, this is only possible when the demand for these products can be estimated, which is nearly impossible at this point in time.

Other limitations can be based on geographical factors (Leon, 2015). It could be possible that there are a lot of banks established in the same region, so in this case in the Malaysian market, but operating in other regions. Since the market shares per region are unknown, the actual market concentration is unknown, since not all banks will operate in every region. This means that the HHI is even less reliable to be used as competition measure. In order to tackle this problem, we should correct for regions and then calculate the HHI again. Due to the fact

(32)

32 that is difficult to obtain information about the regional market shares, it is currently very difficult to conduct this investigation.

7.1.2 Boone indicator

The Boone indicator is of course, like other measures, a simplification of the reality. It requires only information on the profits, in this case the ROA, and on the marginal costs, for which the average total costs are used. Because it specializes in these two variables, it does not take other factors into account. So, it might be that there is a case of omitted variable bias (OVB). This means that a richer dataset, perhaps containing characteristics on the level of the institution, could give an unbiased estimate of the Boone indicator. For example, when a bank is efficient, it could use these gains to invest in order to be able to behave competitively in the future and to cope with possible new entrants to the market which are maybe more innovative (Leon, 2015). These gains are not taken into account since the ROA is calculated using net returns. For example, investing these gains could result in lower costs per unit produced, which also affects profitability of the bank in the next year (Van Leuvensteijn et al., 2011).

Van Leuvensteijn et al. (2011) also mention that this is mostly the case with calculations over the years of the Boone indicator and therefore it is better to use full sample periods. The banking sector may be seen as a market that produces the same products over time, so we could speak of homogeneous products in the banking sector. This is the case in the long run because it is possible that one Islamic bank decides to offer a new Islamic product, so in the short run it will not yet offer homogeneous products.

Another limitation is the fact that there is no benchmark for the Boone indicator. The only known factor is that the more negative the indicator, the more competition there will be in the sector. Therefore, one is able to determine the relative level it is better to be able to compare this to other sectors in the same country. Or maybe also a comparison with the same banking sectors in other countries that have the same structure as Malaysia.

7.2 Alternative measures

There are some other measures that can be used to calculate the competition. These measures will shortly be explained and checked if these methods are applicable to measure the competition of Islamic banks in this paragraph.

(33)

33 7.2.1 Lerner index

The Lerner index is being used to estimate the market power of firms. The price and the marginal costs for the firm are being used to determine market power. If the price and marginal costs are equal, which means that the value is 0, there is perfect competition. The more the values diverge, the greater the monopoly power (Leon, 2015). The equation for the Lerner index is:

𝐿 (𝑞) = S (<) T UV (<)_S (<) (6) As stated before, it is difficult to measure marginal costs of firms, in this case for banks. This means that, just like with the Boone indicator, a proxy has to be used in order to compute the index. This is one limitation of the Lerner index.

But, the Lerner index has an advantage over the use of the HHI. For the Lerner index, you can calculate the market power without having to define the market, which the HHI is failing to take into account. Another advantage of the Lerner index is that fact that you do not need much observations (Leon, 2015). With respect to the Islamic banks, this could be an interesting variable, because the number of Islamic banks in Malaysia is limited. Also, the Lerner index is specific for each bank (Elzinga and Mills, 2011), which allows us to make a comparison of market power between banks. The HHI is only to compute the concentration of the entire market and not the market power of individual banks.

We saw with the HHI that it is not really a proxy for measuring competition, but only for market concentration. The Lerner index faces a similar limitation, since only the market power can be computed, but not the degree of competition in a market.

Another disadvantage of the Lerner index is the fact that it could overestimate market power. Because of the simplicity of the model, the index could be overestimating or underestimating the market power of the firm, since the efficient use of resources is not taken into account (Elzinga and Mills, 2011).

7.2.2 Panzar-Rosse model

The Panzar and Rosse model (P-R) measures the strength of competition based on a reduced form equation (Bikker, Shaffer, & Spierdijk, 2009). P-R explains revenues in terms of input prices of the firm. The equation takes also additional explanatory variables into account (Abel and Roux, 2016). The formula for P-R is as follows:

(34)

34 ln(𝑅𝑒𝑣_$%) = ∝ + ∑+ 𝛽_$%ln (𝜔_$%)

$FG + ∑+\FG𝛾\%ln (𝐶𝐹)+ 𝜀$%, (7)

where Rev states the total revenues, ωi the i-th input factor, CF denotes the other control

variables that capture the bank specific factors. The subscript i stands for bank i, and the t for time. For the input factors, the following ratios are being used: interest expenses to total deposits, personnel expenses to total assets and other operating and administrative expenses to total assets (Abel and Roux, 2016). In order to calculate the value of the P-R model, the H-statistic, we need to take the sum of the input price elasticities:

𝐻 = ∑+ 𝛽_$%

$FG . (8)

The H-statistic varies between 0 and 1. When the value is negative, there is a monopoly situation and when the H-statistic is higher than 1, it indicates perfect competition (Abel and Roux, 2016). The H-statistic can be used together with the Boone indicator as a robustness check, just like Baarsma and Vooren (2018) did. Only, for this thesis it was not possible to calculate this value because there were no personnel expenses and administrative expenses available in the dataset that is used.

Another advantage of the P-R model is that there is no need to specify the market, only the data of the firm is required to calculate the H-statistic. The H-statistic is thus based on the one the data of the individual company, in this case a bank. Also, if there are differences between banks, this can be encountered via the H-statistic (Claessens and Laeven, 2004). Thus, it therefore allows us the make a comparison between different types of banks. This is an effective way to evaluate whether Islamic banks are similar in competitiveness to conventional banks.

A disadvantage of the H statistic is that it could be negative. When the statistic is negative, you cannot reject the fact that there is perfect competition. This makes the statistic questionable (Toolsema, 2002). Toolsema (2002) also says that the statistic is not clear from a theoretical perspective. So, when it is negative we cannot draw the right conclusions about competition.

(35)

35

Chapter 8: Conclusion

The aim for this thesis was to investigate whether Islamic banking has increased competition in the Malaysian banking sector. This is done by using methods that measure the concentration and the competition in the market.

Firstly, the practices of Islamic banking were explained. The way of conducting business is based on Islamic principles. It prohibits receiving and paying interest, which makes the conventional banking products unusable. The important principles in Islamic banking are the prohibitions of riba, maysir, gharar and trade or investment in goods that are haram. The

Sharia boards control if the Islamic products comply with Islamic rule. Because the Muslim

scholars could differ from opinion, there are setting boards, for example the AAOIFI and the IFSB. Islamic banking has its own version for conventional banking products, like mortgages, insurances and loans.

Afterwards, the development of Islamic banking is depicted. This development can be divided in three phases. In order to have a clear view of the development, a timeline of the important milestones in the Islamic banking sector is of Malaysia is provided in figure 7.

Figure (7) Development Malaysian Islamic banking sector

As can be seen, Islamic banking in Malaysia has developed very quickly in a short period of time.

The banking sector in Malaysia is discussed to provide an overview of this sector. This is done by using the SCP model. There are 11 dual institutions in Malaysia. These banks have their own Islamic and conventional bank. Furthermore, there are 5 Islamic banks and 16 conventional banks. A large share of the Malaysian population is Muslim; however, this is not a good proxy for demand for Islamic finance products. The market is regulated by BNM and the MyCC. Because of the strict rules, it is harder to enter the market as a new firm.

1963, Establishment of Tabung Haji 1982, Committee to investigate Islamic banking 1983, Enactment IBA enacted and establishment of BIMB 1993, Interest-free banking Scheme for conventional banks 1994, BNM introduced Islamic Interbank Money Market 1999, Founding of second Islamic bank BMMB 2004, Issuance of three other Islamic banking licenses 2010, Establishment IILM

Competition in Malaysian Islamic banking

UNIVERSITY OF AMSTERDAM

FACULTY OF ECONOMICS AND BUSINESS

Competition in Malaysian Islamic banking

Preface

Table of Contents

Chapter: 1 Introduction

Chapter 2: Literature review

Chapter 3: Islamic banking and its principles

Chapter 4: Development Islamic banking in Malaysia

Chapter 5 Malaysian banking sector

Dual institutions assets (x 1000 USD)

Asset growth, top 10 banks Malaysia (x 1000

USD

)

Chapter 6: Measuring competition

HHI Malaysia

Chapter 7: Discussion

Chapter 8: Conclusion