Abstract
Djibouti: Selected Issues
Financial Inclusion in Djibouti1
A. Summary
The liberalization of Djibouti’s banking sector in 2006 led to an increase in the number of banks, as well as to an improvement in the indicators reflecting access to financial services. The level of financial inclusion, however, remains low, as most of the adult population and small and medium-scale enterprises do not have access to financial services. The financial sector is largely dominated by banks, whose business is concentrated with a few economic players. To broaden financial inclusion, in order to promote inclusive growth with a view to reducing poverty, it will be necessary to strengthen financial infrastructure, develop the microfinance sector, and improve economic formalization.
B. Introduction
1. Access to financial services is essential for inclusive, more equitable growth to foster significant poverty reduction. Demirguc and Klapper (2013) found that, without such access, individuals must use their own resources to invest in their education or to become entrepreneurs, and small-scale enterprises must rely on their limited resources to take advantage of growth opportunities, which promotes income inequalities and impedes growth. Beck, Demirguc-Kunt, and Levine (2007) find that financial development leads to a disproportionate increase in revenue in the poorest quintile, and therefore reduces income inequality and increases the efficiency of the capital allocation which in turn boosts total factor productivity. Hannig and Jansen (2010) uphold that financial inclusion can improve financial stability, specifically by diversifying risks as the result of a more efficient allocation of resources.
2. Despite the rapid expansion of the banking system in recent years, access to financial services remains insufficient in Djibouti. As a result of the liberalization of the banking system, the number of banks has increased from two in 2006 to ten currently. However, the level of banking penetration remains low. Less than 20 percent of Djibouti adults hold a deposit account with a bank and less than 8 percent have access to a loan account in 2014. Outstanding deposit reached 82 percent of GDP whereas outstanding loans from commercial banks, although increased recently, only reached 32 percent. Only 4 percent of adults could obtain a loan from a formal-sector financial institution in 2011, while 18 percent of adults had obtained credit from their circles of family or friends.2 Access to credit is one of the main constraints for formal sector enterprises. Only 5 percent of formal-sector enterprises have access to bank financing. Djibouti lags behind many other countries in the region in terms of number of bank accounts, bank branches and ATMs (Figure 1 and Table 1 in the appendix).
Djibouti: Evolution of Deposit and Credit in the Banking System
Citation: IMF Staff Country Reports 2016, 249; 10.5089/9781498386449.002.A003
Sources: Financial Access Survey, International Monetary Fund, 2014.C. Credit and Deposit in the Banking System
3. Despite a high level of liquidity in the banking system, increased competition between banks, and growing demand for credit, access to financial services, and particularly to bank financing, is still limited for most of individuals and small and medium-scale enterprises (SMEs). The level of bank financing for small and medium-scale enterprises is still quite low, representing only 12 percent of total credit allocated to enterprises. Collateral is required for 84 percent of loans in Djibouti compared to 79 percent in lower middle income countries, and on average 228 percent of the loan value is required as collateral (Figure 4). Certain requirements constrain access to credit, such as the 2 percent fee for the registration of collateral for mortgage loans and another 2 percent fee for the release of collateral. Most SMEs are situated in the informal sector and are excluded, de facto, from access to bank credit as they cannot provide banks with the required documentation (administrative, tax, and financial records) or the required collateral. Almost two thirds of an estimated population of 10,000 enterprises in 2010 operates in the informal sector. Djibouti was ranked 180th out of 189 in the World Bank Doing Business 2015 report in terms of getting credit.
Credit and Deposit in Djibouti and Other Countries
Citation: IMF Staff Country Reports 2016, 249; 10.5089/9781498386449.002.A003
Sources: Financial Access Survey, International Monetary Fund, 2014.Djibouti: Bank Loan Concentration and Risks in the Banking System
Citation: IMF Staff Country Reports 2016, 249; 10.5089/9781498386449.002.A003
Sources: Djibouti authorities; and IMF staff estimates.Djibouti: Number of Accounts versus Credit
Citation: IMF Staff Country Reports 2016, 249; 10.5089/9781498386449.002.A003
Sources: Financial Access Survey, and IMF Staff Calculation.4. In terms of financial depth, Djibouti compares favorably to countries in the region. Credit to GDP and deposits to GDP—which amounted to 32 percent and 82 percent respectively in 2014—are rough indicators of financial depth. However, the high levels of credit and deposits to GDP contrast with the low level of financial inclusion, indicating a high level of concentration among a few individuals and enterprises. Indeed as a result of the banking system liberalization, credit to the private sector has significantly increased from DF 33.9 billion in 2007 to DF 90.3 billion in 2014. The high level of concentration among a few clients leads to exceed the single borrowing limit by all banks.
D. Financial Services Constraints in Djibouti
5. The constraints that limit access to financial services are found on both the supply and demand sides. On the demand side, obstacles to access are connected primarily to absolute poverty at 22 percent of the population, and to unemployment, which affects around half of the active population. Increased competition among banks has led to easier conditions for opening accounts, even though costs of holding and using accounts are still an obstacle, particularly for the low-income sector. The scope of the informal sector leads to widespread use of cash in transactions and limits the use of banking services.
6. On the supply side, while the liberalization of the banking system has led to improved financial intermediation, the latter is still fairly limited. The gap is still substantial between average lending and deposit rates, which are respectively 13 percent and 3 percent. In light of the fairly low levels of inflation and the low cost of funds, the cost of credit substantially reflects banks’ risk premium. Banks apply high lending rates and/or require substantial collateral (see graphs) owing to the significant risk of default associated with small and medium-scale enterprises as a result of information asymmetry, deficiencies in the credit reporting system, lack of credit histories, and the absence of reliable financial statements. As mentioned before, Djibouti banks require, on average, 228 percent of the loan application amount in collateral, compared to 190 percent for the MENA and 179 percent for the SSAs.1
Djibouti: Banks Lending
Citation: IMF Staff Country Reports 2016, 249; 10.5089/9781498386449.002.A003
7. Banks’ capacity to grant medium and long-term financing is constrained by the structure of their resources of which 75 percent are short-term deposits. Two thirds of credit granted by banks is for short-term financing. Moreover, the absence of an interbank market and a lender of last resort require banks to maintain a liquidity surplus to cover their clients’ needs and potential losses. Less than 40 percent of the funds collected are re-injected into the economy in the form of credit, and the remaining is generally placed abroad.
8. The lack of financial infrastructure is another obstacle to access to financial services. The credit information system and payment system are key components of this infrastructure. The 2009–10 Financial Sector Assessment Program (FSAP) identified substantial deficiencies in Djibouti’s financial infrastructure, and recommended that this infrastructure should be strengthened. However, so far progress in this area has been limited.
9. Djibouti’s credit reporting system is comprised of the risk reporting and analysis center and the central register of unpaid checks. While the CBD manages these two sources of information, their scope suffers from substantial limitations. Similarly, the payment systems dominated by cash is still underdeveloped. The establishment of new banks has made it possible to introduce new financial services and instruments, although their range and scope remain limited.
10. The financial sector is characterized by a substantial prevalence of banks and insufficient diversification among institutions and financial instruments. The banking sector is by far the main component of the financial sector, accounting for more than 95 percent of the sector’s total assets and 106 percent of GDP. While the sector is dominated by the two largest banks, representing nearly 67 percent of its total assets, the share of these banks is decreasing as a result of competition from new banks. There is no interbank market.
11. The liberalization of the banking system has promoted the establishment of Islamic banks, targeting customers wishing to comply with Sharia laws. Four Islamic banks were created but the emergence of Islamic banks has not led to a diversified supply of Islamic financial instruments yet.
12. The microfinance sector has remained undeveloped since its emergence in 2008. The Caisse Populaire d’Epargne et de Crédit (CPEC) of Djibouti, the main micro-finance institution, has been under receivership by the central bank due to management problems from 2012 to 2015. A development strategy for the sector was prepared and adopted in 2012, although it has yet to be implemented. Despite its problems, membership in the CPEC of Djibouti exceeded 14,000 in 2014, equivalent to 2–3 percent of the country’s adult population, although the outstanding balance of credit distributed by the institution is still negligible (less than 0.2 percent of bank credit) owing to insufficient financial resources.
13. On the other hand, the Fonds de Développement Economique de Djibouti (FDED) supports the establishment and development of SMEs, primarily by extending credit. For that end, it has access to several credit facilities granted by donors and guaranteed by the government. However, its operating scope is still limited by offering loans between DF 3.5-50 million, and above all by the requirement of joint collateral from a state employee or public enterprise.
14. Djibouti’s tax system is an obstacle for small enterprises intending to operate in the formal sector. Irrespective of profit level, such enterprises are required to pay a minimum lump-sum tax of DF 120,000 from the first year, which represents approximately 10 percent of the average turnover in small-scale enterprises. In addition, Djibouti’s regressive tax system favors large enterprises, as taxes proportion decreases, the larger the enterprise. This situation tends to keep SMEs in the informal sector, limiting their access to bank credit and their development opportunities, while also restricting the expansion of the tax base.
15. The unreliability of accounting and financial information provided by Djibouti SMEs tends to exacerbate the risk perception by banks. On the other hand, banks often lack services specialized in SMEs and procedures adapted to the insufficient financial information and insufficient levels of collateral.
E. Model Description and Calibration Policy
16. Djibouti faces the same obstacles to financial inclusion as many other developing countries. These obstacles can be grouped into three broad categories:
Access barriers are reflected in the high administration cost for opening, closing and maintaining an account. Red tape, guarantee requirement, employment and salary requirements, and deposit requisite decrease the demand for financial services.
Barriers to financial depth are generally measured by collateral requirements, credit and deposit ratios, information disclosure requirement and contract reinforcement.
Intermediation efficiency reflects the supply side barriers by competition among financial institutions and the degree of information asymmetry in the credit market. The inefficiency is then reflected in interest rate spreads and banks’ overhead costs.
17. The analysis of financial deepening effect on growth and inequality is based on Dabla-Norris et al (2015) in which a micro-founded general equilibrium model is developed. The model includes heterogeneous overlapping generation agents which are differentiated from each other by their wealth and talent. Talented and wealthy enough agents choose to become entrepreneurs whereas the non-talented or not-wealthy enough agents remain workers. Two financial regimes are analyzed: in the “saving only” regime, agents can save but cannot borrow. In order for agents in this regime to move to the credit regime, they need to pay a participation cost. Once they are in the credit regime, they are credit constrained by the amount of collateral they can post. The amount of collateral represents the financial depth in the economy. Finally, because of asymmetry of information between lender and borrowers, higher interest rates are charged to cover the monitoring cost and NPL of high leveraged firms.
18. Financial deepening influences economic growth and inequality through many channels. First, a more developed financial system better channels funds into the economy. Second, the channeled funds reach more productive agents in the economy and, where financial costs are high due to financial frictions, the optimal allocation of resources becomes important. Finally, an efficient allocation of financial resources can boost the total factor productivity and economic output. However, high financial deepening can give access to credit to less productive agents which in turn can decrease the TFP and increase the level of non-performing loans. Inequality in this model can also increase or decrease according to the distribution of talents and initial wealth. Indeed, if credit is only allocated to wealthy talented entrepreneurs, income inequality can rise. Therefore it is important that, by lowering the participation costs, poor agents can benefit from financial inclusion and get the chance to move to the credit regime.
19. The model is calibrated using Enterprise Survey and other banking data. The initial wealth for each generation matches the gross saving rate in Djibouti in 2013. Maximum leverage ratio is matched by the median value of collateral divided by loan value using Enterprise Survey (ES). Bank monitoring cost matches the dispersion between interest rate spread and the default rate. This is in fact a wedge between the interest rate spread and the default rate of entrepreneurs because when the monitoring cost is high, the interest rate spread is also high but the default rate is low. Using the ES data, the fixed cost to financial access is estimated by the number of firms with access to credit. The recovery rate after default is determined the judicial system quality and such that the there is a positive default rate. The probability that projects fail is also estimated by interest rate spread. Finally, the agents’ talents are distributed according to a Pareto distribution where the parameter matches the employment distribution.
20. The results of simulation suggest that financial deepening can increase the output, decrease the inequality and improve the TFP.
21. Reducing participation costs: The participation cost is reduced from 0.15 to 0. Lower participation cost in Figure 6, has a positive impact on output and percentage of firms with access to credit. The TFP increases first, which results in a drop in the interest rate spreads and a lower level of non-performing loans. This is also a result of the wealth effect: when entrepreneurs become richer and the participation cost is lower, they ask for less credit and tend to deleverage. However, when all firms finally have access to credit, some of non-productive entrepreneurs start to default, which raises both the non-performing loan level and the spread, and decreases the TFP. Inequality follows the same pattern: when only productive agents have access to finance and participation cost in the financial market is still high for non-productive agents, inequality marginally rises but with a full access for all firms, the Gini coefficient rapidly drops. Therefore, access to finance benefits more the poor population, but it generates higher growth even though the TFP can fall.
Comparative Statistics: Reducing Participation Costs
Citation: IMF Staff Country Reports 2016, 249; 10.5089/9781498386449.002.A003
Source: IMF Staff Calculations22. Relaxing borrowing constrains: Relaxing the borrowing constraint, λ, from 1 to 3 has a positive effect on economic output, productivity and percentage of firms with access to finance. In this exercise, the impact can even be more important than the participation cost effect because although many individuals and firms have loan accounts, they are still subject to credit constraints and cannot borrow from formal financial institutions. But, as the borrowing constrain relaxes, entrepreneurs leverage more which results in a higher interest rate spread and non-performing loans. With higher leverage, the Gini coefficient moves up because talented entrepreneurs leverage more and become richer, but then it stabilizes at certain level since higher spread shrinks entrepreneurs’ profit and wealth.
Comparative Statistics: Relaxing Borrowing Constrains
Citation: IMF Staff Country Reports 2016, 249; 10.5089/9781498386449.002.A003
Source: IMF Staff Calculations.Increasing intermediation efficiency: By reducing the monitoring cost, χ, i.e., increasing the intermediation efficiency, output grows and the interest rate spread decreases. This, in turn, increases firms’ access to finance. However, higher access to finance means that entrepreneurs with a lower productivity can get credit which results in an increase in non-performing loan levels and the decline in the TFP. Two opposing effects are indeed responsible for the decline in the TFP; first, a lower intermediation cost, increases access to finance which boosts entrepreneurs’ wealth and profit. Second, the less productive entrepreneurs can default and increase the NPL levels. In this configuration, the second effect dominated since the TFP has declined. These two opposing effects are also responsible for the step increase in the Gini coefficient.
Comparative Statistics: Increasing Intermediation Efficiency
Citation: IMF Staff Country Reports 2016, 249; 10.5089/9781498386449.002.A003
Source: IMF staff calculationsF. Policy Recommendations
Despite the progress Djibouti has recently made in the area of financial inclusion as a result of expansion in the banking sector, most individuals and SMEs still do not have access to financial services, which limits their capacity to take advantage of economic opportunities created by the substantial investments that have been made or that are now in progress. Accordingly, there is a risk that the scope of the reforms aiming to improve the business climate and the economic effects from investments might be inhibited by insufficient access to financial services. To maximize the economic benefits of these investments and promote inclusive growth, a strategy should be prepared and implemented including the following measures:
23. Development of financial infrastructure
The credit information system should be strengthened. An effective information system that enables real-time access to information on credit applications reduces risks in connection with information asymmetries between lenders and borrowers, and promotes credit to the private sector.
Implementation of the credit guarantee fund for SMEs should be accelerated. The risk sharing principle of the fund should encourage banks to grant loans to SMEs and promote formalization.
Payment system reform is required to promote innovation. Mobile telephone service, which covers more than half of the adult population, can be used to accelerate financial inclusion by offering financial services more suited to the public’s needs at a lower cost than traditional banking services.1
Diversification of financial instruments should be promoted through implementation of an adequate framework that supports the emergence of new financial mechanisms such as leasing.
24. Microfinance
Implementation of the microfinance strategy, adopted in 2012, should be accelerated, and collaboration should be encouraged between banks and microfinance institutions to offset the lack of resources in microfinance institutions, enabling them to obtain refinancing from banks.
25. Formalization of the informal sector and support for SMEs
Tax system reform should offer incentives to SMEs in the informal sector to move towards the formal sector, in order to give them greater access to financial services, and specifically to bank credit.
Facilitated access to property ownership and improved mortgage procedures can promote access to credit by allowing the use of land as collateral, which would reduce financial risks for banks.
26. Promoting greater access to financial services
Reducing the participation cost in financial markets, relaxing borrowing constraints and increasing the financial intermediation efficiency could increase growth and reduce inequality. However, as shown in the simulations, different policies can have different impacts in term of growth and inequality.
Lowering costs of opening bank accounts and better access to credit are key elements of a global financial inclusion strategy. A lower participation cost not only has positive impacts on growth, but it can also decrease the interest rate spread and reduce inequality and non-performing loan levels. Relaxing borrowing constraint can increase the economic growth although the financial system would be more exposed to risks. Increasing the intermediation efficiency also boosts the growth and reduces the interest rate spread. In this regard, bank supervision must be strengthened to ensure the stability of the financial system. An extended financial inclusion strategy can make the financial system more stable through risk diversification by reducing credit concentration risks.
Appendix I. Access to and Use of Financial Services
| Access to & Use of Financial Services | |
|---|---|
| 2014 | |
| ATMs per 100,000 adults, Number | 7.31 |
| ATMs per 1,000 km2, Number | 1.86 |
| Commercial bank branches per 100,000 adults, Number | 5.61 |
| Commercial bank branches per 1,000 km2, Number | 1.42 |
| Deposit accounts with commercial banks per 1,000 adults | 159.77 |
| Loan accounts with commercial banks per 1,000 adults | 79.78 |
| Outstanding deposits with commercial banks as percent of GDP | 82.57 |
| Outstanding loans from commercial banks as percent of GDP | 32.77 |
| Access to & Use of Financial Services | |
|---|---|
| 2014 | |
| ATMs per 100,000 adults, Number | 7.31 |
| ATMs per 1,000 km2, Number | 1.86 |
| Commercial bank branches per 100,000 adults, Number | 5.61 |
| Commercial bank branches per 1,000 km2, Number | 1.42 |
| Deposit accounts with commercial banks per 1,000 adults | 159.77 |
| Loan accounts with commercial banks per 1,000 adults | 79.78 |
| Outstanding deposits with commercial banks as percent of GDP | 82.57 |
| Outstanding loans from commercial banks as percent of GDP | 32.77 |
References
T. Beck, A. Demirguc-Kunt, and R. Levine, 2007, Finance, inequality and the poor.
Dabla-Norris, Era, and others, 2014, “Financial Deepening, Growth, and Inequality: A Structural Framework for Developing Countries,” IMF Working Paper (forthcoming) (Washington: International Monetary Fund).
A. Demirguc-Kunt and L. Klapper, 2012, Financial Inclusion in Africa, Policy Research Working Paper No. 6088, June African Development Bank, 2015, Djibouti Country Strategy Paper, 2013-15,.
International Monetary Fund, 2009, Djibouti Financial system assessment program.
A. Hannig, and S. Jansen, 2010, Financial Inclusion and Financial Stability: Current Policy Issues, Asian Development Bank Institute, Working Paper No. 259 (December).
