UNIVERSITY OF AMSTERDAM
AMSTERDAM BUSINESS SCHOOL
Master in International Finance
FOREIGN BANK ENTRY AND FINANCIAL INCLUSION IN AFRICA
By
Suso Camara Fatou
Program: MIF
Student Number: 10880801
Thesis Supervisor: Prof. Stefan Arping
August 2015
Contact details of the author: fsuso83@hotmail.com
TABLE OF CONTENTS FOREIGN BANK ENTRY AND FINANCIAL INCLUSION IN AFRICA ... 1 TABLE OF CONTENTS ... 2 LIST OF FIGURES ... 3 LIST OF TABLES ... 3 1. INTRODUCTION ... 5 2. LITERATURE REVIEW ... 8 2.1. WHY DO FOREIGN BANKS EXIST? ... 9
2.1.1. FINANCIAL SYSTEM ENVIRONMENTAL FACTORS ... 9
2.1.2. BANK‐SPECIFIC FACTORS ... 10
2.2. THE FORM AND THE MODE OF FOREIGN BANK ENTRY. ... 11
2.3. FINANCIAL INCLUSION AND FOREIGN BANKS ENTRY ... 12
3. DATA AND METHODOLOGY ... 16
3.1. THE HYPOTHESIS ... 18
4. DESCRIPTIVE ANALYSIS ... 19
5. MAIN RESULTS ... 28
5.1. REGRESSION RESULTS ... 28
5.1.1. FOREIGN BANKS AND FINANCIAL INCLUSION ... 28
5.1.1.1. HETEROGENEOUS EFFECTS ACROSS COUNTRIES ... 29
5.1.2. FOREIGN BANK ASSETS AND FINANCIAL ACCESS ... 30
5.1.3. FOREIGN BANKS AND COST OF FINANCIAL SERVICE ... 32 5.1.4. FOREIGN BANKS AND FINANCIAL SECTOR STABILITY ... 34 6. CONCLUSION ... 35 REFERENCES ... 36
List Of Figures
FIGURE 1 FOREIGN BANKS AS PERCENTAGE OF DOMESTIC BANKS, BY COUNTRY ... 20
FIGURE 2 FOREIGN BANK ASSETS AS PERCENTAGE OF TOTAL BANK ASSETS, BY COUNTRY ... 21
FIGURE 3 FOREIGN BANK ASSETS AND FINANCIAL INCLUSION ... 22
FIGURE 4 FOREIGN BANK ASSETS, INDIVIDUAL BANK ACCOUNTS AND GDP PERFORMANCE ... 23
FIGURE 5 FOREIGN BANK ENTRY OVER TIME, AND THE GLOBAL FINANCIAL CRISIS ... 24
FIGURE 6 AFRICA' PROGRESS IN FINANCIAL SECTOR DEVELOPMENT OVER TIME ... 25
FIGURE 7 GDP PERFORMANCE AND FINANCIAL ACCESS ... 26
FIGURE 8 FINANCIAL INCLUSION, THE POOR AND THE ILLITERATE... 27
List of Tables TABLE 1 LIST OF SOME KEY VARIABLES USED IN THE ANALYSIS ... 18
TABLE 2 FOREIGN BANKS AND FINANCIAL INCLUSION (ACCOUNTS PER 1000 ADULTS) ... 29
TABLE 3 EFFECT OF FOREIGN BANKS ON FINANCIAL INCLUSION (NUMBER OF ACCOUNTS) IN POOR AND NON‐POOR COUNTRIES ... 30
TABLE 4 FOREIGN BANK ENTRY AND ATMS PER 1000 ADULTS ... 31
TABLE 5 FOREIGN BANKS AND OVERALL BANKING SECTOR COST EFFICIENCY ... 33
TABLE 6 FOREIGN BANK AND THE FINANCIAL STABILITY (NONPERFORMING LOANS) ... 34
ABSTRACT
In this paper, we attempt testing four related but distinct hypothesis of the effects of foreign bank entry in a domestic financial system. Using data from the World Bank’s financial database (FINDEX, 2015), we find heterogeneous effects of foreign banks in African countries. For countries that are relatively wealthy, with mean per capita income above the sample average, foreign bank entry has resulted in significant financial inclusion. For less wealthier countries with per capita income less than the sample mean, entry of foreign banks has fail to achieve substantive financial inclusion. For an average country however, foreign bank entry is found to be associated with lower cost of financial intermediation and more financial stability as proxy by interest margins and nonperforming loans respectively. The findings suggest that Africa needs to address financial exclusion using both demand-side and supply-side policy interventions. Further liberalizing the financial sector can potentially achieve significant service expansion through lower cost and more geographical distribution of financial services.
1. INTRODUCTION
Financial inclusion has attracted a lot of attention among governments and development agencies around the world. Central banks, international Monetary Fund, World Bank and other financial development and regulatory institutions have dedicated enormous resources towards achieving more inclusive financial systems. In effect, promotion of financial inclusion has become a fundamental development objective in many countries. It is believed that financial inclusion promotes employment, sustainable economic growth and most importantly, bringing the rest of the unbanked population into the financial service arena.
In the past decades, many countries in Africa have experienced a large inflow of foreign banks. Therefore it is important for policy makers to closely look at the impact of these bank entries on expansion of financial services to the larger unbanked population. In simple term, Financial Inclusion can be explained by the proportion of a country’s population that has access to financial service. Access to financial service can be explained by two related factors. First, a country may have low level of financial inclusion because most people in the country cannot afford financial services. On the other hand a country’s low level of financial inclusion may be explain by the absence of adequate financial service provision, even when the population can afford it. These two distinctions can be summarized by “affordability” and “accessibility”, respectively. In this context, we evaluate Financial Inclusion and how it has been influenced by the rate of foreign bank entry in various African financial markets. Given that Africa as a continent exhibit huge heterogeneity both in terms of average income levels and the developmental state of financial sectors, the presence and effect of foreign banks in different countries may not have homogenous effects on the level of Financial Inclusion in these countries. We therefore, try to evaluate whether the effect of foreign bank entry on the level of financial inclusiveness is dependent on these heterogeneous factors.
Since banks are the main providers of financial services in most African countries, then the focus of this paper will be on how the entry of foreign banks meets the need of the larger population that were previously having no access prior to entry of foreign banks. In most
analyses that evaluate the effect of foreign bank entry on financial Inclusion, the evidence has shown mixed results. In a handful of findings, the results indicate that increased foreign bank presence result in expansion of the financial sector, resulting in extended access to those who were previous excluded. In another set of findings, more foreign bank entry has often not resulted in any meaningful increment in the level financial inclusion. In both sets of findings however, the approach has commonly being in the framework of cross country regression which lumps developed and developing countries into a single data set in evaluating the hypothesis. These aggregated cross‐country regressions miss important peculiarities, especially among developing countries that are characterized by both demand‐side constrains as on the side of the clientele, and supply‐side constrains on the side of the financial service providers. For example, the presence of multiple financial service providers in a developed country may directly explain the level of financial service uptake. Whiles on the contrary, the presence of the same level of financial services may not directly translate into the same financial service uptake. The simple reason for these differences may be that in most developing countries, where majority of people live on less than one‐dollar a day, it is often difficult to device financial services or products that are responsive to the needs of these extreme poor groups. Under such circumstance low level financial inclusion is explain by demand‐side factors such as poverty. For one to do meaningful understand the effect of foreign bank entry, especially in Africa, it may be necessary to evaluate this hypothesis, conditioned on both supply‐side and demand‐side factors.
Giving the scope of this study, contrary to previous studies, limits itself to evaluating the effect of foreign bank on financial inclusion among African countries, we try to put special emphasis in understanding and explaining factors that make the relationship between foreign bank entry on Financial Inclusion differ between African countries rather than between African and non‐ African countries. The paper critically looks at financial inclusion in three aspects: the number of bank accounts, Number of branches and finally the number of ATM. In addition, it looks at cost of financial intermediation and financial stability proxy by repayment performance of the sector.
A number of studies made previously in this area have shown that Foreign Bank entry has a positive impact on the overall development of a country’s financial sector. The reason for this is that foreign bank entry leads to competition and as a result, it pushes banks beyond their traditional comfort zone into providing services to customers that are previously having no access to financial services.
The rest of the sections will be as follows: Section 2 will look at literature review. Section 3 Data and methodology, Sections 4 descriptive section 5 Regression results and finally section 6 will give the conclusion.
2. LITERATURE REVIEW
The main purpose of banks coming into existence is financial intermediation. Before the coming of banks, lenders found it difficult to give out their money because of lack of trust in customers and risk of default as well as lack of information about borrowers. On the other hand, borrowers who are in need of money to finance their businesses and to take care of other immediate financial needs find it difficult to get lenders who will trust them and are willing to trust them with their money. Even if there are few people willing to lend, the ability to pool large amount of funds and finance bigger investments was often not feasible. Therefore banks came in to fill this gap between the lenders and borrowers.
The primary role of the bank is to intermediate between lenders and borrowers. In other words, lenders give their money to banks to do business with it, which includes lending to borrowers. The bank pays an interest on these funds at a regular interval as a reward to lenders. The bank in turn lends this money to borrowers at an interest usually higher than what they pay to lenders. They also take additional collateral to secure the safety of lenders money in case of a default. Others fees are usually charge by the bank for doing these services. The difference in the two interest rate and other fees is what the bank use as their gain. Larger transactions are often more favorable for banks as costs associated with such are relatively smaller when compared to cost of doing several small transactions. The cost saving associated with large transactions is called economies of scale. Often banks are selective on the type of clients they deal with in order to enjoy economies of scale. For this reason, large fractions of populations are left with no access to services of the bank due to their low incomes. Development policies attempting to address the lack of financial services for these people have looked into possibilities of designing financial products that require less transaction cost. In many cases, these products are unattractive to commercial banks but are a market for non‐ bank financial institutions such as microfinance institutions.
of financial service in a country. In the latter set of restrictions, increased number of financial institutions combined with good policies can lead a more developed financial system. The ability for banks in minimizing transaction cost has made them crucial intermediaries in the financial sector. By extension, one can say that banks are very important for economic growth because they reduce market imperfection.
2.1. Why do foreign banks exist?
The reason why foreign banks enter domestic markets can be divided into two. First, environment‐specific reasons in financial system and secondly, banks‐specific reasons. Various environmental factors such as political, cultural and religious, geographical location and rate of return divergence may be source of attraction for foreign banks. On bank‐specific factors, strategic positioning to certain markets and the need to diversify investments may trigger their movements to foreign markets.
2.1.1. Financial System Environmental factors
A country’s political stability can encourage the inflow of international banks. If a country is politically stable it strengthens and also promotes relationship between other countries. This in turn can encourage foreign bank inflow. Another important factor that influences foreign bank existence is the culture, language as well as the religion in a country. Countries that have the same culture and language are likely to conduct more trade amongst them. Religious factor is also an important influential factor. For instance, Islamic banks operate more in countries were Islam is the dominant religion. This may be because of the difference in banking service and products provided by Islamic banks. For example, Islamic bank does not encourage the use interest and this can be better understood by Muslims than other people belonging to other religion.
Another area of interest to the foreign banks is the interest margin, if the interest margin in a country is wide, foreign banks with potentially higher profitability can take advantage and penetrate into the system by providing competitive rates and thus increasing their shares. Finally, geographical location: Banks from the same region are likely to trade.
2.1.2. Bank‐Specific Factors
In some countries, banks can follow their customers if their customers have businesses in other countries. They help those customers in payment of remittance in their home countries. Also domestic banks may be hesitant to extend credit to foreign customers, but if they have their home country banks, that can give them easy access to credit. One important advantage for banks to follow their customers is that the customers can help them get information about local businesses and other local customers. However foreign banks that follow their customers are also interested in host country customers to expand their customer base to earn economies of scale.
The lack of strict regulation in a financial system can increase or encourage the inflow of foreign banks. If banks realized weak spots in a countries regulation, they tend to take advantage and penetrate that system and make their gains. (Fieleke, 1977; Goldberg and Saunders, 1981) concludes that banks tend to move from one location to another if they realize a regulatory gap in the system where they intend to move to. Countries can avoid this by imposing stricter and more uniform measure of bank supervision and regulations. In a situation where stricter regulations are imposed on foreign banks, domestic banks can have a comparative advantage over them.
The most important benefit of international diversification is to minimize risk. Therefore banks extending their businesses by physical presence in other countries can reduce their risk significantly. For instance if there is a problem in one economy the presence in the other country can reduce of neutralize the overall risk that the bank will face. Even though this might involve high cost in the short term, the long‐term benefits often outweigh the short‐term costs. Intangible assets like “good will” is important for banks for profit maximization and increasing market share. Banks can expand to other foreign location to attract customers because customers will have confidence in them due to the reputation they already gain in their home countries. Therefore banks find it importance to use this type of asset to expand in foreign financial system ( Hefferman 1996).
2.2. The form and the mode of Foreign bank entry.
There are different ways foreign banks can enter into a foreign location. They can enter as an agency, branch or a subsidiary.
Agency: foreign banks can operate as an agency in foreign location if they only want to do a
simple banking business like loan arrangement and other fund transfers. This is a very redistricted and simple form of banking services that banks can choose to engage in. If a bank is not sure of what a foreign economy looks like, they can start operating as agency in that country until they have clear information about the economy (Clarke et al., 2001).
Branches: if a bank intends to expand their business and engage in broader banking business,
they can operate as a branch. Most often, banks do not need a full banking license in the host country to operate a branch. Branches are supervised by home country regulatory authorities. However, host country’s authorities also participate fully in all their supervision. A branch is not required to submit their regulatory returns to the host country regulators instead they consolidate it with the home country reports.
Subsidiaries: subsidiaries operate with a banking license from the host country. They are
required to go through all the licensing process like domestic banks. Operations are fully supervised and regulated by host country. However in some circumstances there can be a joint supervision by home and host country. Financial statements and all other regulatory reports are submitted to the host country regulators for reviews. Therefore we can conclude that subsidiaries have exactly the same operations and regulations as domestic banks (Clarke et al., 2001). There are other forms which foreign bank can enter into a foreign market. Foreign banks can enter into foreign markets through acquisition or joint venture. This will save the banks from going through all the bureaucracy of the licensing processes by host country depending on how much shares they want to acquire. One advantage of foreign bank entry through acquisition is that the bank can save a lot of cost for information gathering because the bank that they are taking over already knows the market.
Joint venture is also another form of entry. This form of entry is important for host economy because it restrict the bank from owing a major share. Entry in the form of joint venture has the same advantage as acquisition. Access to information will be easy as the partner domestic bank will already have good knowledge of the local economy and the financial system at large.
2.3. Financial Inclusion and Foreign Banks Entry
There is a large literature on the importance of financial inclusion and a host of factors that influence a country’s level financial inclusion. While some researcher think financial inclusion is improved by high foreign bank presence in a financial system, others think it can hamper growth of the domestic banking sector and suppress sustainable development of the financial the system. We therefore discussed the findings by different researchers in the rest of literature.
Beck et al. (2007) has found that financial inclusion and the expansion of financial services is important because it gives opportunity to poorer household and also small business that may not have adequate collateral to get credit to be able to have access to formal credit market. Beck et al also stated that the reason why we have such people left out in the financial service could be as a result of market imperfections. Such imperfection can be as a result of high cost of banking transaction, which makes it difficult for poorer people to afford and also make gains from their investment. Information asymmetries and lack of proper legal system are also a contributing factor. This type of market imperfection can hurt economic growth because there may not be fair distribution of resources and the poor will stay poorer.
Karen Ellis (June 2007) finds that financial inclusion has impacted positively by foreign bank entry because it improves small firms access to credit indirectly. The fact that they put pressure through expansion and loan provision in the urban areas force other banks to move from urban to rural areas, and thus providing credit to smaller firms who were not having access before. Another advantage of foreign bank presence was by (Levine 2001). In his research, he found that the presence of foreign banks leads to improvement in financial infrastructure through development of rating agencies, credit reference bureaus, regulatory and supervisory
institutions as well as auditors. Another important research by (Goldberg 2007) highlights the association between increased numbers of foreign bank and connected lending.
There are numerous other advantages associated with foreign bank entry to a country’s financial system: Improvement in the efficiency of the financial system is an important benefit often attributed to entry of foreign banks. These banks come with advance technologies which make banking services much easier thus reducing cost and improving efficiency. This puts pressure on domestic banks as they may not want to loose market shares to the foreign bank due to slow and inefficiency service. As a result, foreign bank entry leads to an overall improvement in the efficiency of countries’ financial services. (Fries and Taci 2005), in their paper, argue that generally, foreign banks are consider being more efficient than domestic bank. The coming of these banks in a country’s financial system leads to transfer of technology as well as other human expertise to the domestic banks. Foreign banks come with greater innovation in a system by the provision of new banking products to customers. The increase in financial stability has also been an important effect of foreign banks. These banks have the advantage of international diversification thus reducing the overall risk of a country. They also have a better and effective risk management system in place than domestic banks.
Access to foreign capital is another important benefit of foreign bank. In addition to their ability to increase competition and improve supervisory reforms, foreign banks also helps a country to have access to foreign capital (Levine 1996). Some researchers find that foreign banks follow their customer. However, (Clarke et al., 2001) argue that the opposite can happen. Both ways can be beneficial to a country because it helps a country have access to foreign capital.
Finally, the issue of foreign bank seen as “cherry pickers” was looked at from a different perspective. Because they are careful in their loan disbursement, that makes them have quality and less volatile loans compare to domestic banks.
Whiles all the above discussions revealed a positive relationship between foreign bank entry in a financial system and financial development, several other studies have found a negative link between the two. Sarma and Pais (2008) conduct a cross country analyses to determine the relationship between financial inclusion and development. They found that a financial system
that is said to be inclusive promotes both economic growths by bringing the larger portion of the population into the financial system. Thus this helps them adopt saving practices and also gives them opportunity to access other financial service. They also emphasis that in other to have a proper financial development, it has to move in apparel with human development although there may exist some exceptions. However, looking at their paper in the context of our paper, Sarma and Pais (2008) have found a negative correlation between foreign bank participation in a financial system and financial inclusion. Ellis (2007) has researched into whether foreign bank entry can have positive impact on financial inclusion. She noted that if foreign bank entry is followed by high concentration of banks, it can deviate from this purpose and result into a negative financial inclusion. A study by Clarke et al. (2006); Beck and Brown (2014) and Gormley, (2010) look at the behavior of foreign banks in terms of lending to the poor. They found that foreign banks tend to be involved in “cherry picking” behavior. This means that they only lend to client that are involve in transparent businesses. The fact that clients involve in small and opaque business are left out leads to a reduction in the number of people having access to credit. Detragiache et al. (2008) as well as Beck et al. (2007) both did a research on foreign bank entry and financial outreach and they both found similar result of a negative correlation between the two. Detragiache et al. (2008) conducted his work on 60 low income countries and found that high presence of foreign leads to a decline credit. This is so because foreign bank tend to be involve in “cream skimming” as a result smaller and more opaque firms tend to be left out in credit accessibility. In a recent case study done by Beck and Martinez Peria (2010) conducted on Mexico, they conducted a study on the behavior of municipalities where there is a high participation of foreign bank and found that actual high presence of foreign bank leads to a reduction in number of bank accounts, branches as well as bank loans.
Foreign banks can also affect an underdeveloped economy by involving in capital flight. In most cases profit made by this banks are not reinvested in the economy instead they send this out in a form of different payment to their parent bank. Capital flight by foreign banks can harm a country’s economy especially during economic crises because the funds needed to improve the
Finally, Cull and Martinez Peria (2007) Conduct a research on the timing of foreign entry into a country’s financial system and found out that high foreign bank entry can lead to an increase in credit. However if the entries come after a crises, they can have negative correlation with private credit. This is because most foreign bank takes over an already financial distress bank. In this case they conclude thus foreign banks are important in stabilizing financial system. Another study conducted on 1600 banks mainly in developing countries from 1996‐2002 concluded that the reason that resulted to a reduction in private credit by foreign bank is that these banks have access to liquidity from their parents so they don’t have flexibility to make decisions in giving credit. As a result informal and opaque small firms find it difficult to access credit from them (Mian, 2003).
The argument still continues as to whether foreign bank entry as more advantages or disadvantages to financial inclusion and development especially in developing financial economies. Therefore there is a need for countries to carefully evaluate the issue before taking decisions as to whether foreign banks should be totally stop from entry into other economy’s or freely allowed to expand their services to foreign markets. Some countries have strict restriction on foreign bank entries. For instance, in a country like Egypt, Foreign banks can only operate in a form of joint venture with a local bank (Caprio and Cull, 2000).
3. DATA AND METHODOLOGY
The data we use comes from the World Bank financial data including the detailed financial sector information on 53 African countries. For each country, we capture the level of foreign bank presence in two possible ways. First, we capture foreign bank by the number of foreign bank as percentage of total banks. This measure gives us an estimate in terms of a count but not in terms of how large the investment of foreign banks is relative to domestic banks. Secondly, we measure foreign bank presence in terms of the percentage of banking sector assets that belong to foreign bank as a proportion of total banking sector asset. In contrast to the first measure, this second measure depicts the level of investment of foreign bank relative to domestic banks. It is difficult however to conclude weather more large number of foreign bank or large investment by foreign bank is a better indicator of their influence on the level of Financial Inclusion. Giving that both measures are use in the literature, we try to test the relationship between foreign bank entries on financial inclusion using these two indicators for foreign banks.
First, we look at the effect of foreign bank entry on the number bank accounts as a measure of financial inclusion in a country. While domestic or nationalized banks are more likely to be influenced by governments towards opening accounts for public sector institutions, I belief that evaluating the effect of foreign banks on number of bank accounts in general is more unbiased way of evaluating the effect of foreign bank entry in a domestic economy. Second outcome variable is the number of bank branches or ATM outlets. These are indicators of the geographical distribution of financial service provision within a country. Also, the more widely distributed bank branches are and the denser this distribution implies better access to the financial service in that country. In this regard, I look at the effect of foreign bank entry on branch and on ATM distribution. Looking at the banking outreach, meaning access to physical service points, the World Bank global financial inclusion index (“findex”) survey reveal that 2.5 billion worldwide are without access to financial services. These people are said to be excluded
bank accounts is the distances to service point. In other for policy makers to achieve their objective of financial inclusion, provision of access point to the greater population should be improved. This has made bank branches a very important variable in my analyses of the impact of foreign bank entry bank branches and ATM in Africa.
Physical access to point of service has been a major obstacle of not having formal bank account for many people (World Bank, 2014). In most countries, a number of reasons lead to banks inability to provide physical point of services to larger population especially to those living outside the urban areas. In addressing the financial access concern, we first use the Geographical distribution of banking services measured by number of bank branches per 100,000 of he population in every African country between 1990 and 2011. Our second measure of financial access is the number of ATMs per 100,000 adult of a country’s population. We do not assume that these two measures capture the same effect. One of the most important reasons is that services points such as establishing branches can come with high cost to the banks. However, ATMs can serve as a better substitute for this because they are much more cost effective compare to branches. A third assessment looks at the effect of foreign bank entry on cost of financial intermediation in the economies. We evaluate the effect of foreign bank entry on the entire banking sector’s interest and cost margins, including interest spread, noninterest income and overhead cost. We hypothesized that through competition effects, foreign banks are likely to reduce a country’s cost of financial intermediation. More foreign banks lower interest spread in the domestic financial sector, an outcome that is good for consumers of financial services as it imply that the interest margin that banks make on deposits are relatively small. In another case, we look at the effect of foreign banks on banks' non‐interest income and on overhead costs. A lowering effect in each of these implies a better outcome for consumers of financial services.
Our fourth assessment looks at a measure of financial instability as proxy by the level of a banking sector’s nonperforming loans. In this case, out attempt is to understand whether foreign banks contribute to the overall health of the financial system in terms of the quality of loans and the rigor in loan recovery.
3.1. The Hypothesis
This paper attempts to test four related but independent hypothesis that links foreign bank entry to various outcomes of financial inclusion, expansion, service cost and stability.
Hypothesis 1: Increase in foreign bank entries in African countries has resulted to an increase in financial inclusiveness (the number of bank account)
Hypothesis 2: increase in foreign bank entries in Africa lowers cost of financial
intermediation (interest spread, noninterest income and overhead costs)
Hypothesis 3: Increase in foreign Bank resulted to an increase in the distribution of
financial access points across a country. E.g. ATM services, bank branches etc. Hypothesis 4: Increase in foreign bank entry resulted to an increase in financial stability (nonperforming Loans) In Table 1 below, we provide definition of some key variables used in this paper. Table 1 List of some Key Variables used in the Analysis Variable Definitions No. Foreign Banks Number of foreign banks as fraction of total banks in the country; Foreign Bank Asset Fraction of banking sector assets belonging to foreign banks; ATMs per 100,000 Number of Automated Teller Machines per 100 thousand of the population; Bank acc. per 1,000 Bank accounts per 1,000 adults Population Total population of a country in a given year; NBFI Assets in GDP NBFI Assets in GDP“is Nonbank financial institutions’ assets to GDP (%); Remittance to GDP (%) Remittance inflows to GDP (%); Non performing loans Bank nonperforming loans to gross loans (%) GDP Per Capita Gross domestic per capita in constant 2000 USD Fin. Sys deposit to GDP (%) Total financial system deposits to GDP (%) Bank deposit to GDP (%) Total banking sector deposits to GDP (%) Middle‐income country
A countries between 33th and 66th percentiles of mean income distribution of African countries High‐income country A countries above the 66th percentiles of mean income distribution of African countries Interest spread Bank lending‐deposit interest spread Noninterest income Noninterest income as fraction of total income (%)
4. DESCRIPTIVE ANALYSIS
This section presents summary statistics based on the key variables used in this paper. The graph bellows shows that there is high presence of foreign bank in most of the 53 African countries used in our analyses. The graph shows that a significant number of countries about (50%) of those with data available have much less than 50% presence of foreign banks. While in some countries the dominant domestic banking sector is characterized by private domestic ownership, a number of countries still have significant government presence in the domestic financial sector. As hypothesized, one will expect relatively less influence on the financial sector in state dominated financial markets, especially with limited foreign bank entry. In most cases, privately owned domestic banks often find it difficult to compete with state banks. Similarly, state banks enjoy the patronage of the larger government institutions, which often make a significant market share of banks’ clientele in developing countries. The second graph shows the total assets own by foreign bank to total assets of the banking industry.
Figure 1 Foreign banks as percentage of domestic banks, by country
Another way of measuring the extent of foreign bank presence in an economy is to estimate the fraction of total banking sector assets that are owned by foreign banks. While the mare number of foreign banks as fraction of total banks can be misleading in the sense that there may be several small foreign banks relative to domestic banks, the use of bank assets is more likely to reflect the true extent of foreign bank presence in an economy. Larger bank assets in addition to measuring financial inclusiveness, is also indicative of developmental stage of the economy. More advance economies relative to less developed ones are likely to have larger bank assets. In the chart below 17 out of 33 listed African countries, foreign bank assets exceed
half of all banking sector assets. This means that foreign banks do not only dominate developing country markets by theirs numbers not also in terms of their investment.
Figure 2 Foreign bank assets as percentage of total bank assets, by country
To further understand the relation between foreign banks entry and domestic financial inclusiveness indicators, we split African countries into two sub‐groups based on share of total bank assets belonging to foreign banks. Countries with more than 50 of the banking sector assets owned by foreign banks are categorized as high foreign bank presence while those with domestic banking sector dominance are considered low foreign bank presence. Based on these two groups, we compare countries performance in some financial inclusion indicators. First, in the graph below, both the number of bank branches per thousand of the population and the number of ATMs per thousand of the population seem to be higher among countries with relatively less foreign bank dominance. This finding seems to suggest that foreign banks lead to
lower fin that driv For exam dominan the cont number politician Figure 3 Source: W Similarly, with less banks. In to bank a banks, th 3000 peo 0 1 2 3 4 5 6 7 8 coun more (Ban nancial inclus e the associ mple, it coul nce is limited rary, countr of bank bra ns may have Foreign ban World Bank , bank acco concentrati n this figure, accounts in c he number o ople per 10 t ntries with e FB assets Cou les nk branches pe sion. It is im iation betwe d be the cas d and more ries with hig
nches opera kept them o nk assets and FINDEX (201 unts per 10 ion of foreig while less th countries wh of bank acco thousand of untries with ss FB assets co m er 100,000) (A mportant to een foreign se that fore market driv gh governme ated by gov operational d financial in 15) Note: FB ,000 people gn bank com han 1500 pe here more th ounts in coun f the populat ountries with more FB assets ATMs per 100, note howev bank presen ign banks a en factors d ent presenc ernment. So for other no nclusion B stands for f e of the pop mpared to tho eople per 10 han 50 per c ntries with le tion. This m Countries with less FB assets ,000 adults) ver that ther nce and the re entering determine fin
e in the ban ome of thes onprofit driv foreign bank pulation is r ose with hig thousand o cent of the b ess foreign b eans people h s (Bank b countri (Bank b Countr (ATMs countri (ATMs Countr re can be so financial in markets wh nancial serv nking sector se may not b en factors. k. relatively hig gher concent of the popula bank assets b bank presen e in countrie branches per 1 ies with more branches per 1 ries with less FB per 100,000 ad ies with more per 100,000 ad ries with less FB ome other fa clusion mea here govern vice provision r may have be profitabl
gher in coun tration of fo ation have a belong to fo nce is in exce es with less f 100,000) FB assets 100,000) B assets dults) FB assets dults) B assets actors asure. ment n. On large e but ntries oreign ccess oreign ess of fewer
foreign b countries When on graph be The mea than me foreign b Figure 4 Source: W For one t financial and also financial the mea observed African e presence 0 500 1000 1500 2000 2500 3000 3500 banks tend s with more ne looks at th elow shows n per capita an per capit banks are mo foreign ban World Bank to effectivel inclusion, it within cou sector varia sure of for d that both economies h e and their t countries wit more FB asset
Bank acco
to be abou foreign ban he type of e that foreign a income of e ta income o ore prevalen k assets, ind FINDEX (201 y evaluate t t is appropr ntries over ables have b eign bank p the number have been h total asset i h ts Countries w FB ass ounts per 10,00 ut 50% less ks. conomies w n bank tend economies w of economie nt in poorer A dividual ban 15) Note: FB the effect of iate to obse time. In the been changin presence ha r of foreign igh in years nvestment i with less sets cou mo 00 likely to ha where foreign to dominat with high fo es with low f
African econ nk accounts a
B stands for f f foreign ban erve what fa e next sectio ng among A as changed banks and t s prior to 20 in Africa plu untries with re FB assets GDP pe ave bank ac n banks are e in econom reign bank p foreign ban nomies and GDP pe foreign bank nks on the d actors have on, we try t African econo
in the cont the total am 007. The sud ummeted be Countries with FB assets r capita ccounts rela more likely mies that are presence is 2 k presence. erformance k. domestic eco changed be to look at h omies. First, tinent over mount of ass
dden decline etween 200 h less
s
ative to tho
to dominate e relatively 25 percent l This shows onomies’ lev etween coun how fundam , we look at time. It ca sets they ho e in foreign 7 and 2009 ose in e, the poor. ower s that vel of ntries mental t how an be old in bank . This
downtur impact h institutio develope below. H recover f of foreign explained alternativ investme plus the recovery Figure 5 Source: W 0 10 20 30 40 50 60 70 FY n can be at highest felt ons operatin ed countries However, du from the 200 n financial in d by the fa ve investme ent in more resilience from the gl Foreign ban World Bank Y2004 FY200 ttributed to t in the wo ng in Africa , the African ue to their l
07 global fin nstitutions a act that Afr ent destina developed c shown by obal financia nk entry ove FINDEX (201 05 FY2006 the financia orld’s must and other d n financial se ess integrat ancial shock and their inv rica and so tion for mo countries. Th most Africa al crisis. r time, and 15) Note: FB FY2007
al crisis that t developed developing c ector has su ted nature, k. The recove estments in me other d ost investor his diversific an economie the global fi B stands for f FY2008 FY2
t hit the glo d countries. countries ty uffered the s most count ery is eviden Africa. The developing rs that had cation towar es contribu inancial cris foreign bank 2009 FB FB obal financia . Because f ypically origi shock as can tries in Afric nt by the inc
sudden reco countries w d highly co rd less conta
ted to the
sis
k.
B assets % of to B number % to
al sector wit foreign fina inate from t n be on the g ca were qui creasing pres overy can al were becam ncentrated agious econo continent’s otal tal th its ancial these graph ck to sence so be me an their omies s fast
Overtime 2004 and distributi more peo the popu thousand noted th its histor with the Figure 6 Source: W In the fig people w economi income p 0 2 4 6 8 10 12 e, Africa as a d 2009 subst ion and spr ople that it ulation incre d of the pop at this perfo rical perform developed w Africa' prog World Bank gure below, with access t es. The num per capita of ATMs per 100 a whole has tantial finan reading ban previously u eased for 2 t pulation incr ormance is i mance. On th world, revea gress in finan FINDEX (201 Africa’s fina o bank acco mber of ban f the populat ,000 adults s registered ncial sector e k branches used to be th to as high a eased from impressive w he other ha als significan ncial sector d 15) Note: FB ancial sector ounts and it h nk accounts tion has stea Bank bran impressive f expansion as meant that he case. Num s 10. In the 3 in 2004 to when one co nd, compari t disparities developmen B stands for f r expansion has been un per hundre adily grown ches per 100,0 financials se s indicated b t financial p mber of ATM same way, o about 5 in ompared Afr ing Africa w in financial nt over time foreign bank is also refle nderpinned b ed thousand between 20 000 FY FY FY FY FY FY ector develo by a steadily provision ha Ms per hund bank branc n 2009. It sh rica’s recent with the rest sector deve e k. cted in incre by s slow bu d of the pop 004 ad 2009. Y2004 Y2005 Y2006 Y2007 Y2008 Y2009 opment. Betw y increasing as reached m dred thousa ches per hun ould howev t performan of the wor elopment. eased numb ut steady gro pulation, and . ween ATM many nd of ndred er be nce to ld, or ber of owing d the
Figure 7 Source: W Finally, t benefits excluded poorest 4 the popu educated the rich h same vei that disp remained will prov and effec Bank acc GDP perform World Bank to examine of financia d. To do this 40% of the ulation in A d people dis have made s n, financial a parities betw d wide. Add ide cost effe ctively utilize ounts per 10,0 GDP per cap mance and F FINDEX (201 the true ex al sector de s, we look a population Africa. In the
sproportiona significant g access was n ween the ric dressing this ective servic e. 0 500 000 pita Financial Ac 15) Note: FB tent of fina evelopment at the over t
and compar e same way ately enjoye ains in term not limited t ch and the gaps will n ce while rem 1000 1500 20 ccess B stands for f ncial inclusi has occurr time trend i re it to finan y, we comp ed benefits. s of financia to educated poor and b eed innovat maining less 000 2500 3000 foreign bank ion, one ne red to the
in financial ncial access
are if educa Evidence sh al access bet people. It is between the tive financia complex for 0 3500 FY FY FY FY FY FY k.
ed to exam poor who access betw made by th ated people how that bo tween 2011 s important t e literate an al product d r the majori Y2009 Y2008 Y2007 Y2006 Y2005 Y2004 mine whethe were previ ween the bo he richest 60 e relative to oth the poor and 2014. I to note how nd illiterate evelopment ty to unders er the ously ottom 0% of o less r and n the wever, have t that stand
Figure 8 Source: W Account primary Account sec financial inc World Bank Account (% Account fin p Account fin ri at a financial in y education or at a financial in condary educa clusion, the FINDEX (201 0 % age 15+) poorest 40 chest 60% nstitution, r less (%) nstitution, tion poor and th 15) Note: FB 10 he illiterate B stands for f 20 30 foreign bank 40 k. YR2014 YR2011
5. MAIN RESULTS
To determine the true relation between foreign bank entry and financial sector development, a mare correlation tells us little about this relationship. Because financial inclusion or financial sector development is affected by so many other variables other than the level of foreign bank presence in a domestic economy, controlling for those measurable factors that are likely to contribute in determination of a country’s financial inclusion is pertinent. To do such analysis, we implement a multiple regression analysis that accounts for the partial effect of several other factors on the extent of financial inclusiveness.
It is important to note that while we have made efforts to ensure that our regression specifications include as many relevant variables as possible, data limitation and other problems inherent in establishing causality remain outstanding. We therefore do not claim causality in the results that are presented below.
5.1. Regression Results
In the results section, we discuss the effects of foreign banks on financial inclusion measured by number of accounts per 1000 adults; on financial access measured by number of ATMs per 100,000 adults, and the number of bank branches per 100,000 adults; on cost of financial intermediation measured by Interest spread, noninterest income and overhead cost; and on financial stability measured by banking sector nonperforming loans.
5.1.1. Foreign Banks and Financial Inclusion
The results below shows that overall, within Africa, there are evidence that increased presence of foreign banks have contributed to the expansion of financial services. Increasing the number of foreign banks as fraction of total banks by one percent is associated with an increase of about 5 new accounts per 1000 adults the population. The results also show that a country's population and the number of non‐bank financial institutions are associated with fewer accounts per thousand of the population. On the other hand, more remittance flows into an African country and the level of the country's development are associated with more financial inclusion.
In the analysis below, we evaluate the effects of foreign bank entry in poor and non‐poor African countries. In the data, we distinguish between these two sets of countries by using the Per capita GDP (in constant 2000 US Dollars). The mean per capita GDP in the data is USD1533. We define all countries with means capita GDP above the threshold as non‐poor while those with mean per capita GDP less than the threshold are categorized poor.
Table 2 Foreign Banks and Financial Inclusion (Accounts per 1000 Adults) BankAcc per
1000 Adults
Bank Acc per 1000 Adults
No. Foreign Banks 4.843** 4.843**
(1.785) (1.785) NBFI Assets in GDP -32.11** -32.11** (11.91) (11.91) Population -0.00000308 -0.00000308 (0.00000197) (0.00000197) Remittance to GDP (%) 66.99*** 66.99*** (12.51) (12.51) GDP Per Capita 0.0369*** 0.0369*** (0.00752) (0.00752) Constant 6.852 6.852 (102.1) (102.1) R-sq. 0.847 0.847 N 25 25 Standard errors in parentheses * p<0.10, ** p<0.05, *** p<0.01 note: African countries only 5.1.1.1. Heterogeneous effects across countries In the first set of results, the dependent variable is number of bank accounts per 1000 of the population and the main independent variable of interest is the fraction of banks that are foreign in a country. The results show that the effect of foreign banks on the number of bank accounts per 1000 of the population positive and significant. That means foreign banks contribute to financial expansion among the population. This effect is true for both poor and
rich countries. However, it can be seen that the effect is much larger for rich countries. While a one per cent increase in the number of foreign banks only increase number of accounts by approximately 1 account per 1000 adults, it results in as large as 14 new accounts per 1000 adults for countries that are categorized non‐poor. This suggests that there are other factors that explain the low level financial inclusion in poor countries.
Table 3 Effect of Foreign Banks on financial inclusion (number of accounts) in Poor and Non‐ Poor Countries
(Accounts per 1000 Adults)
Non-poor Countries
Poor Countries
No. Foreign Banks 13.94** 0.831*
(5.063) (0.363) NBFI Assets in GDP -29.46*** -20.16** (6.836) (4.838) Population -0.0000135*** -0.0000842*** (0.00000356) (0.0000129) Remittance to GDP (%) 46.79** 12.40 (19.68) (7.192) GDP Per Capita 0.0111 6.276*** (0.0133) (0.898) Constant -69.54 -176.3*** (359.5) (33.32) R-sq. 0.938 0.992 N 15 10 Standard errors in parentheses * p<0.10, ** p<0.05, *** p<0.01. Note: African countries only 5.1.2. Foreign Bank Assets and financial access
Moving from merely considering foreign bank presence as measured by their number among total banks to a more substantive measure that accounts for their level of investment in the domestic economy, we present another specification to explain the effect of foreign banks. In this case, assets owned by foreign banks as fraction of total banking sector assets measure foreign bank presence. Preliminary results from this regression show that foreign bank assets
have not resulted in significant financial inclusiveness in Africa. The coefficient to our foreign bank variable is negative and significant. Table 4 Foreign Bank entry and ATMs Per 1000 Adults ATMs per 100,000 ATMs per 100,000 ATMs per 100,000 Linear Interaction
Foreign Bank Asset -0.0203 0.0338 0.0338
(0.0254) (0.0380) (0.0380)
Fin. Sys deposit to GDP
(%) -4.101*** -3.511*** -3.511***
(0.720) (0.551) (0.551)
Bank deposit to GDP (%) 4.409*** 3.744*** 3.744***
(0.731) (0.560) (0.560)
Credit to private sector
to GDP (%) 0.213*** 0.204*** 0.204***
(0.0245) (0.0187) (0.0187)
Bank’s return to Equity 0.0980*** 0.0838*** 0.0838***
(0.0342) (0.0267) (0.0267) Liquid Liabilities -0.000129*** -0.0000914*** -0.0000914*** (0.0000252) (0.0000217) (0.0000217) Middle-income country 8.025** 8.025** (3.702) (3.702) High-income country 4.935 4.935 (3.974) (3.974) Middle-income*Foreign Bank Asset -0.0808* -0.0808* (0.0478) (0.0478) High-income*Foreign Bank Asset 0.0932* 0.0932* (0.0512) (0.0512) Constant -5.249*** -11.40*** -11.40*** (1.966) (3.379) (3.379) R-sq. 0.804 0.892 0.892 N 107 107 107 Standard errors in parentheses * p<0.10, ** p<0.05, *** p<0.01 note: African countries only When one evaluates the effect of foreign bank assets on level of financial inclusion in Africa, with a more disaggregated data that caters for heterogeneity among African countries, it can be seen that foreign banks have an effect that is dependent of the level of development of individual countries. The association is unclear. The coefficient to the foreign bank variable is
insignificant. On the other hand, when one distinguishes between poor and relatively more advanced African countries, the effect turns different.
For a set of countries with mean capita above the continent's mean, increase in foreign banks is associated with more financial inclusion. This provides another validation of our hypothesis that in poor countries, demand‐side factors such as ability to afford financial services is the main factor that limits financial inclusion. In more developed countries, supply side constraints such as availability of financial service providers tend to be the key constraint.
In the regression results above, an interaction between the foreign bank variable and the level of development shows that poor countries, relative to non‐poor countries are less likely to experience more financial inclusion with entry of foreign banks. It shows that the standalone effect of foreign bank entry is positively associated with ATMs per 100,000 of the population. The interaction shows however, for poor countries, the net effect is negative and statistically significant.
Similarly, if one group African countries into three categories based on per capita incomes, the least affluent countries are more associated with non‐improving financial indicators despite large foreign bank entry. In the final specification (column 2 Table 4), countries with per capita GDP in excess of USD927 are likely to have 1 more ATMs per 100 thousand of the population with every ten percent increase in assets owned by foreign banks.
Along side out main independent variable, we have seen that other control variables such as the willingness and ability to deposit among customers, domestic credit demand by private sector and profitability of the banking sector (banks' return on equity) are important determinants of financial expansion.
5.1.3. Foreign Banks and cost of financial service
This part of the results looks at the effect of foreign banks on cost of financial service. Specifically, in Table 5, we evaluate the effect of foreign bank entry on the entire banking sector’s interest and cost margins, including spread, net interest, noninterest income and
spread in the domestic financial sector. This outcome is good for consumers of financial services as it imply that the interest margin that banks make on deposits are relatively small. In another specification, we look at the effect of foreign banks on banks' non‐interest income. The results show that financial systems with more concentration of foreign banks are associated with lower non‐interest charges. A 10 percent increase in number of foreign banks lowers noninterest charges by about 3 percentage points. In the third specification, we can see that bank overhead costs are relatively lower in countries where banking sector is dominated with foreign banks. Table 5 Foreign banks and Overall banking Sector cost efficiency Interest Spread Noninterest Income Overhead Costs
No. Foreign Banks -0.0562* -0.299* -0.0807***
(0.0298) (0.137) (0.0199)
GDP (Current 2000 USD) -2.31e-11* (1.37e-11) credit to deposit ratio -0.0908*** (0.0268)
Bank return on Asset -2.504*
(1.506) Bank return on Asset(after tax) 2.090 -12.69*** 0.761*** (1.326) (3.573) (0.185) Banks ‘Assets to GDP (%) -0.0986*** (0.0330) ATMs Per 100,000 2.509** (1.015)
Bank Acc Per 1000 Adults -0.0925* 0.0121***
(0.0415) (0.00316) Working Capital -0.735 (0.706) Bank deposit to GDP (%) -0.263*** (0.0473) Constant 24.05*** 103.2*** 14.85*** (3.246) (21.18) (1.876) R-sq. 0.251 0.649 0.466 N 200 13 87 Standard errors in parentheses * p<0.10, ** p<0.05, *** p<0.01 note: African countries only
5.1.4. Foreign Banks and financial sector stability
Part three of the results focus on assessing the effect of foreign banks on tendency of financial crisis (bank crisis) as explained by the rising level of nonperforming loans. In a final hypothesis that we test in this thesis, we look at whether the piling nonperforming loans in African financial sectors are explained by foreign bank presence in local economies. The logic in this argument could be that foreign banks lack personal ties and general knowledge of market risk and are therefore more likely to lead to significant increase in NPLs. Quite in the contrary, the regression results show that changing a typical country's financial sector from mainly domestic to entirely foreign bank players is associated with a lower amount of bad loans in that country by 12 percentage points. It also shows that overall banking sector assets and the level of a country's development are important in averting incidence of bad loans and thus reducing the risk of potential financial crises.
Table 6 Foreign Bank and the financial Stability (Nonperforming loans)
Nonperforming loans
No. Foreign Banks -.129423***
(.0263) Bank Assets -.0032625 (.0358) GDP Per Capita -.0018397*** (.0002) Constant 22.91342*** (3.165) N 135 Standard errors in parentheses * p<0.10, ** p<0.05, *** p<0.01 note: African countries only We are cautious in the interpretation of this result as a causal effect of foreign banks on the country’s financial stability for two reasons. First, the results may be picking the effect that foreign banks tend to enter markets that are more stable and thus have lower nonperforming loans. Secondly, it maybe that both the presence of foreign banks and the level of financial stability in a country are explained by a third factor such as the political stability or other
CONCLUSION
In this paper, we analyze the role of foreign banks in the development of the domestic financial sector and how these effects translate into better financial inclusiveness. We use panel data of 53 African countries covering detailed financial sector records between 1990 and 2011 from the World Bank’s Financial Institutions dataset (FINDEX 2015). We evaluate the effect of foreign banks entry as measured by their number and share in domestic banking sector investments on a number of financial sector outcomes, including inclusion, cost efficiency and stability outcomes. Our findings suggest that heterogeneity among African countries is an important factor in explaining the effect of foreign banks in domestic financial systems. In relatively poorer African countries, financial inclusion tends to be explained by low affordability rather than lack of access. In those economies, foreign bank entry does not lead to any significant expansion in financial services. We explain this as demand‐side factor. On the other hand, in relatively affluent economies, more foreign bank entry is associated with significant improvement in domestic financial inclusive indicators.
Our results also show that regardless of the level of development, foreign bank entry if associated with improved cost efficiency in the financial sector. Interest spreads, non‐interest income and overhead costs are significantly lower in countries with relatively large foreign bank presence. In addition, nonperforming loans are found to be much lower in these countries. Finally, based on the findings here and in much of the literature, policy makers should face the challenge of addressing low‐level financial inclusion both from the demand and supply sides. Building the poor income to have capacity to afford financial services is likely to be as good as making financial services providers extend services to those that are currently marginalized. In this way, African countries are more likely to impact meaningfully on the lives of the large populations that are currently excluded from financial services.
