From 1989 to 1991 Ghana implemented a financial sector reform program aimed at rehabilitating the country’s financial system, which suffered from inefficiency, lack of competition, and a large portfolio of nonperforming loans. Measures were taken to liberalize interest rates, revamp financial sector legislation, and strengthen bank supervision. These reforms were costly. Nonperforming loans were swapped for interest-bearing bonds, and banks were recapitalized, at a cost estimated at about 6 percent of GDP (Dziobek and Pazarbašioğlu, 1997). This section analyzes the strategy used to restructure the Ghanaian banking system and draws some lessons. The section begins by reviewing the historical development of the financial system in Ghana, the early reforms of 1989–91, and the second wave of reforms, characterized by the divestiture process that began in 1992 and is still ongoing. Next the present status of the banking sector is described and compared with the situation in other countries. The section concludes by summarizing the main findings and identifying issues that remain to be addressed.
The Banking System and Its Evolution
The first commercial banks set up in Ghana at the beginning of the 20th century still operate today as Standard Chartered Bank of Ghana and Barclays Bank Ghana (Table 8.1). Under colonial rule their main business was trade finance, and they mainly served the expatriate community. In 1953 the first indigenous commercial bank, now called Ghana Commercial Bank (GCB), opened to provide credit services to the local population. Following independence in 1957, the Bank of Ghana was established to serve as a central bank for the economy and to take over some of the functions previously carried out by the West African Currency Board. Throughout the 1957–83 period, the government followed a policy of intervention in economic activity and held a controlling interest in all commercial banks. Three state-owned development banks were established during this period: the National Investment Bank (NIB), the Agricultural Development Bank (ADB), and the Bank for Housing and Construction (BHC). In 1983, after the launching of the Economic Recovery Program, financial sector reforms were initiated, and private commercial banks were allowed to operate.
Chronology of the Banking System
Chronology of the Banking System
| Date | Event | |
|---|---|---|
| 1888 | Government Savings Bank is set up. | |
| 1896 | British Bank of West Africa (later the Standard Chartered Bank of Ghana) begins operations. | |
| 1912 | West African Currency Board is set up among Gambia. Ghana, Nigeria, and Sierra Leone. | |
| 1917 | Colonial Bank (later Barclays Bank of Ghana) begins operations. | |
| 1953 | Bank of the Gold Coast (later the Ghana Commercial Bank) begins operations. | |
| 1957 | Bank of Ghana is set up. | |
| 1964 | National Investment Sank is created. | |
| 1965 | Agricultural Development Bank is created. | |
| 1972 | Bank for Housing and Construction is created. | |
| 1972 | New government repudiates loans contracted by previous governments. | |
| 1977 | Social Security Bank is set up, wholly owned by SSNIT. | |
| 1981 | Government freezes deposits of 50,000 cedis or more. | |
| 1988 | Private banks are authorized to operate. | |
| 1987 | September | Commercial bank interest rates are liberalized. |
| 1989 | July | Bank restructuring plan is adopted by the government. |
| August | New banking law strengthening the regulatory environment and the supervisory authority of the Bank of Ghana is adopted. | |
| 1990 | January | New management and boards of directors for all financially distressed banks are appointed. |
| April | Implementation of restructuring plans for three financially distressed banks begins. | |
| May | Nonperforming loans to state enterprises by three financially distressed banks are replaced by Bank of Ghana bonds. | |
| November | Requirement that banks lend for agriculture is abolished, and all bank charges and fees are decontrolled. | |
| November | Implementation of restructuring plans for three additional financially distressed banks begins. | |
| December | Nonperforming loans of the private sector in financially distressed banks are replaced, primarily with Bank of Ghana bonds. | |
| December | First compliance test based on new capital adequacy requirements is conducted. | |
| 1991 | March | Restructuring plan for a seventh financially distressed bank is finalized. |
Chronology of the Banking System
| Date | Event | |
|---|---|---|
| 1888 | Government Savings Bank is set up. | |
| 1896 | British Bank of West Africa (later the Standard Chartered Bank of Ghana) begins operations. | |
| 1912 | West African Currency Board is set up among Gambia. Ghana, Nigeria, and Sierra Leone. | |
| 1917 | Colonial Bank (later Barclays Bank of Ghana) begins operations. | |
| 1953 | Bank of the Gold Coast (later the Ghana Commercial Bank) begins operations. | |
| 1957 | Bank of Ghana is set up. | |
| 1964 | National Investment Sank is created. | |
| 1965 | Agricultural Development Bank is created. | |
| 1972 | Bank for Housing and Construction is created. | |
| 1972 | New government repudiates loans contracted by previous governments. | |
| 1977 | Social Security Bank is set up, wholly owned by SSNIT. | |
| 1981 | Government freezes deposits of 50,000 cedis or more. | |
| 1988 | Private banks are authorized to operate. | |
| 1987 | September | Commercial bank interest rates are liberalized. |
| 1989 | July | Bank restructuring plan is adopted by the government. |
| August | New banking law strengthening the regulatory environment and the supervisory authority of the Bank of Ghana is adopted. | |
| 1990 | January | New management and boards of directors for all financially distressed banks are appointed. |
| April | Implementation of restructuring plans for three financially distressed banks begins. | |
| May | Nonperforming loans to state enterprises by three financially distressed banks are replaced by Bank of Ghana bonds. | |
| November | Requirement that banks lend for agriculture is abolished, and all bank charges and fees are decontrolled. | |
| November | Implementation of restructuring plans for three additional financially distressed banks begins. | |
| December | Nonperforming loans of the private sector in financially distressed banks are replaced, primarily with Bank of Ghana bonds. | |
| December | First compliance test based on new capital adequacy requirements is conducted. | |
| 1991 | March | Restructuring plan for a seventh financially distressed bank is finalized. |
In the early 1980s, after decades of state management, Ghana’s banking system was in distress. Banks suffered from undue political influence, weak management, inadequate capital, backward information and accounting systems, and poor internal controls. Moreover, they had large portfolios of nonperforming loans and insufficient provisions against losses, and they were often overexposed to a few clients, in particular other state enterprises. At the end of 1989 the banking system was close to a crisis, as nonperforming loans reached 41 percent of total credit. Faced with this situation, the government adopted a reform program with the objective of restructuring the banking system while enhancing its competitiveness and efficiency. A new banking law was enacted that year, which laid out the basic regulatory framework for the banking system: minimum capital requirements, capital adequacy ratios, prudential lending ratios, exposure limits, and accounting and auditing regulations. The supervisory activities of the Bank of Ghana were also strengthened, and banks were required to submit accounts for off-site monitoring. Annual on-site inspections, as well as off-site surveillance, were to be conducted to verify compliance with regulations.
During 1990 and 1991 most nonperforming loans on the banks’ balance sheets were swapped for government-guaranteed, interest-bearing bonds issued by the Bank of Ghana or were offset against liabilities to the government or the central bank.29 A total of ¢62 billion in nonperforming loans was removed from banks’ portfolios, of which ¢47 billion was replaced with bonds paying 7, 9, and 15 percent annual interest at two- to five-year maturities; these bonds have since been rolled over. A nonperforming assets recovery trust (NPART) was formed in 1990. Its performance as evaluated in 1995 seemed impressive: ¢13 billion was recovered out of a total ¢18 billion outstanding. This assessment, however, does not take into account that these were years of high inflation, which meant that the amounts eventually collected represented much smaller values in real terms.
In February 1992 the government announced a strategy to divest its shares in commercial banks. The divestiture was intended to increase competition and efficiency in the system. The program made early progress but quickly stalled. The Social Security Bank and the National Savings and Credit Bank were restructured and merged: 21 percent of the merged bank’s shares were then divested through a public offer in March 1995, and 40 percent were sold to a strategic investor (Table 8.1). In October 1995, 60 percent of the merged bank’s shares were listed on the Ghana Stock Exchange.
In February 1996, 30 percent of the shares of GCB, the largest bank targeted for divestiture, were floated. After the initial offer was oversubscribed, the government decided to increase the public offer to 42 percent of shares while looking for a strategic investor for 40 percent (the government retained 18 percent of the shares). Substantial delays were encountered in the negotiations with the strategic investor, partly because in 1997 the U.K. supervisory authorities objected to the proposed ownership transfer of the London branch of GCB. Instead, the London branch was converted into an independent bank, Ghana International Bank.30 Discussions with a strategic partner were interrupted in 1999 when the investor withdrew, and the process had to be initialed anew. Discussions with a new partner were still ongoing at the end of July 2000. In June 1998 the government divested three-quarters of the remaining 40 percent of its shares in Barclays Bank Ghana.
The A-life Fraud
In January 1997 the banking supervisory authorities uncovered a large check fraud involving three public banks, the owner of the A-life supermarket chain, and 10 other, associated companies. A-life had opened a series of accounts in three public banks–GCB, BHC, and CooP Bank–both in their Accra headquarters and in their regional branches. Many of the accounts were also opened in the name of the associated companies. Taking advantage of the lengthy clearing time between Accra and the provinces, and with the complicity of the employees and some of the directors of the banks involved, checks were written on accounts held in the regional branches and then cashed in Accra. The checks were held for a long time before they were presented to the clearinghouse and the corresponding accounts debited. The size of the float increased in the clearinghouse, but it reached a considerable amount before the authorities realized the extent of the irregularities and intervened. At that time the total amount of checks in suspense accounts amounted to ¢129 billion, or approximately 1 percent of GDP in 1996. About ¢50 billion worth of checks had been written and cashed under the check purchase facility; the remainder consisted of overdrawn accounts held by the banks involved.
The Serious Fraud Office intervened and froze the assets of the perpetrators; meanwhile the Bank of Ghana banned the purchase of checks over the counter. GCB’s exposure to the fraud was about ¢22 billion, of which 70 percent was in the form of checks purchased. The remainder was written against accounts held in BHC and CooP Bank in equal parts. As a result of the losses incurred by these banks in connection with the fraud, their net worth became negative. Top management of the banks involved was dismissed and imprisoned.
After some investigation and a petition in the High Court, in September 1997 the accounts of the A-life company were unfrozen, and the company was allowed to resume its full business under court supervision, in the hope that the profits generated would help recover some of the debt. A special account was set up to collect proceeds from the sales of the assets and the profits from the commercial activity toward the recovery of the debt. As of May 1998. ¢25 billion had been recovered, but subsequent amounts recovered were negligible. Most of the funds had been transferred abroad before the fraud was uncovered. The trial of the owner of A-life is still pending as of this writing.
Four other banks–NIB, ADB, BHC, and Cooperative Bank (CooP)–had been identified for divestiture at the onset of the program. Discussions had begun in 1995 with a strategic investor for NIB but failed to come to a conclusion. Bids were called twice more, the more recent of which took place in August 1998. Finally, 60 percent of NIB was sold to a consortium of foreign banks in January 2000. The divestiture of BHC and CooP stalled in early 1997 after the A-life check fraud caused significant losses to both banks (Box 8.1), leaving them with negative capital adequacy ratios during the period 1998–99. In January 2000 the government liquidated both BHC and CooP, and their deposits were transferred to ADB. The estimated cost of the operation was about ¢60 billion, substantially less than initially envisaged, and the government is expected to issue special one-year notes to finance this operation.
Structure of the Banking System
The banking system in Ghana at the end of 1999 consisted mainly of 17 banks; 9 commercial banks, 4 development banks, and 4 merchant banks. Altogether these banks account for about 90 percent of deposits in the banking system. The financial sector also includes 107 rural banks (Box 8.2) and several non-bank institutions, including 4 brokerage companies, 2 discount houses, 7 savings and loans institutions, the Social Security and National Insurance Trust (Box 8.3), and some other, minor financial institutions. In January 2000 two more commercial banks were licensed to operate, and two state-owned banks, both of them development banks, were liquidated.
Ghana’s banking penetration ratio is relatively high for Sub-Saharan Africa. On average, Ghana has a bank branch for every 54,000 persons. The Greater Accra region has an average of one branch per 15,000, but density in the north is much lower, with only one branch for every 85,000 persons. About half of the banks have branches in the interior of the country, but GCB alone owns about half of all local branches and is represented in all regions of the country.
Rural Finance in Ghana
Historically, lending to the agricultural sector was mostly carried out by the Agricultural Development Bank, which was set up specifically for this purpose in 1965. In the mid-1970s rural banks were established, primarily to institutionalize financial intermediation in rural areas and mobilize rural savings for on-lending. At the end of 1999 there were 111 rural banks located in nine regions; their distribution is somewhat biased toward the urbanized Ashanti and central regions.
Rural banks are incorporated as limited liability companies. Until 1994 they were owned jointly by people in the catchment area and the Bank of Ghana. Since then the Bank of Ghana no longer participates in the ownership of newly created banks, but it retains shares in banks created before 1994, In general, ownership of rural banks is not confined to the communities where they are located. Rural banks have to operate in conformity with the Banking Act as amended in 1989, and they are subject to Bank of Ghana supervision just as are the commercial banks.
Financial intermediation by rural banks has been increasing considerably in recent years. However, growth in outstanding credit to the private sector has been weak, partly because of the considerable volume of external project assistance, and partly because of the high interest rates available on treasury bills in recent periods. An analysis of the distribution of loans reveals that only 13–16 percent of rural bank credit is directed to agriculture; the greater part is allocated to salaried workers in the urban areas.
Although rural banks still account for only 3 percent of deposits within the formal banking sector, they have been relatively successful in mobilizing rural deposits, mostly in the form of savings and time deposits. The total number of deposits grew more than 25 percent between 1995 and 1997, and in particular they include deposits of informal deposit takers, such as susu collectors, who play an important role in bridging informal and formal entities. (Susu is an informal saving scheme whereby participants agree to pay a certain sum at regular intervals for a specified period of time. In an emergency the depositor can request that the money be paid out earlier than agreed.)
The main problems currently facing the development of rural banking stem from the isolation of the rural communities. These problems include, in particular, poor management skills, inadequate oversight of the supervisory bodies, low staff qualifications, and weak internal controls. Communication and infrastructure problems have limited the services banks can offer, such as check clearing, and this in turn has constrained the adoption of technology and contributed to the high transactions costs. In addition, rural banks have very weak legal instruments for debt recovery, and this undermines confidence in the system. In early 1999, 23 rural banks in distressed financial condition because of nonperforming loans had their licenses withdrawn by the Bank of Ghana.
An Association of Rural Banks was founded in 1981 to serve as an advocate for rural banks and to provide supporting services, including technical assistance. However, the association has no legal authority to enforce its recommendations. Most of the activities of the association are expected to be taken over by a new apex institution.
In 1994 a World Bank study suggested the need for such an institution to provide financial, managerial, and technical support to the rural banks (World Bank, 1994). The apex institution is expected to become operational in November 2000. In particular, it is expected to help rural banks test new products for serving rural clients, disseminate best practices in rural microfinance, and assist rural banks in dealing better with risks. The institution will have a banking license and will be owned by members of the rural banks. It will represent all rural banks as a single entity in the national clearing system, thus reducing the cost to individual banks of participating in the system. In this way the apex institution will be able to provide check clearing facilities to the participating banks, reducing liquidity problems and transactions costs. In addition, the apex structure should foster the development of reporting standards and evaluation criteria and therefore improve official supervision of the rural banks.
Rural banks have great potential to contribute to rural growth and poverty reduction, especially given that more than half of GDP is generated in the agricultural sector. However, their potential is now stifled by the lack of management capacity, inability to adopt new technologies, and other constraints. Key steps to strengthen the sector are the implementation of the apex initiative and the deepening of the links between the rural sector and the informal sector.
Although the number of financial institutions in Ghana is high (and has grown considerably since the early 1990s), financial intermediation remains relatively modest. In 1997 the ratio of M2 to GDP was 17 percent,31 not too different from other countries in Sub-Saharan Africa, but only about 30 percent of the level in most industrial countries (Figure 8.1, top left panel). This ratio changed very little between 1980 and 1999, showing little progress toward financial deepening.32 Holdings of currency are relatively high about 6 percent of GDP, but virtually unchanged since 1990; indeed, they have been declining (Figure 8.1,) top right panel). Credit to the private sector as a share of GDP, after growing sluggishly during most of the 1990s, increased rapidly in 1998 and 1999, to reach 14 percent, three times its level at the beginning of the reforms. However, actual levels remain well below the average for Sub-Saharan Africa (Figure 8.1, bottom left panel). The M1/M2 ratio has shown a moderate decline in the last decade, a development worthy of note in a country prone to inflation (Figure 8.1, bottom right panel). In summary, Ghana seems to be moving toward increasing the depth of its financial system, albeit at a slow pace.
Financial Deepening Ratios
(In percent)
Sources: IMF International Financial Statistics, various issues; Ghanaian authorities; IMF staff estimates1 Data for Sub-Saharan Africa are unweighted averages.Financial Deepening Ratios
(In percent)
Sources: IMF International Financial Statistics, various issues; Ghanaian authorities; IMF staff estimates1 Data for Sub-Saharan Africa are unweighted averages.Financial Deepening Ratios
(In percent)
Sources: IMF International Financial Statistics, various issues; Ghanaian authorities; IMF staff estimates1 Data for Sub-Saharan Africa are unweighted averages.The Social Security and National Insurance Trust
The Social Security and National Insurance Trust (SSNIT) is the government agency with responsibility for collecting social insurance contributions from enterprises and making social insurance payments to retired and disabled workers and their dependents and survivors. From 1965 through 1990 Ghana’s retirement pension system operated as a provident fund: in 1991 it shifted to a social insurance system. The fundamental purpose of this change was to shift Ghana’s social security system from one that pays only a lump-sum benefit to one that pays monthly benefits to members throughout their retirement. The law allowed for wider coverage, opening participation to all employees in the formal and the informal sectors, including self-employed persons who opt to join. The law provides a basic defined-benefit pension for retired workers, benefits for survivors of workers, and disability benefits.
Ghana’s social insurance program is subject to actuarial evaluation every year by the Actuarial Department of SSNIT and once every three years by an independent consulting actuary. These reviews have projected that the trust’s total expenditure will increase rapidly from 5.3 percent of insured earnings in 1995 to 9,8 percent in 2005, which is still well below the current contribution rate of 17.5 percent. These reviews did not recommend a reduction in the contribution rate, because the additional funds will be needed as the cost of the system continues to increase rapidly.
The projected increase in SSNIT’s reserve fund to ¢5.7 trillion at the end of 2005, equivalent to 12 percent of projected GDP, clearly indicates its dominant role in Ghana’s economic development. SSNIT currently has 46 percent of its capital in fixed-income assets and the remainder in non-fixed-income investments. Fixed-income investments, in particular treasury bills, have provided high yields, but yields on non-fixed-income assets, especially real estate and nonlisted equities, have been disappointing. No unlisted equity has yet paid dividends, whereas 14 of 16 listed companies in which SSNIT holds equity have done so. As a result, SSNIT is shifting its policy on asset allocation to one based more on an evaluation of the returns on equity than on the development needs of the country.
SSNIT’s reserve fund represents virtually the only domestic source of long-term funding in Ghana. In 1998 SSNIT’s total assets amounted to about 42 percent of banking system assets, and its inflow of funds was ¢253 billion, approximately 55 percent of the banking system’s increase in deposits. Therefore SSNIT’s investment decisions, in addition to being important for its own financial soundness, are important for the functioning of the financial system. The size and potential influence of SSNIT’s reserve fund in Ghana’s financial markets require a careful balancing of the need to preserve SSNIT’s independence from government interference and the need to ensure that SSNIT does not use its influence to interfere with the normal functioning of markets.
SSNIT is in transition from a provident fund requiring full funding to a social insurance system requiring a lower degree of funding. Further, because pensions only started to be awarded in July 1991, the number of pensioners is low, and expenditures are well below levels typical of a mature program. Therefore, despite its current actuarial soundness, SSNIT faces significant challenges as the largest source of financial capital in an environment with limited investment opportunities and increasing expenditures.
Financial activity remains concentrated in a small number of core banks, although the changes from the late 1980s are more visible in this area. The potential for competition is well above the average for Sub-Saharan Africa, as none of the five largest banks is excessively dominant. The Herfindahl concentration index for these five banks is 0.14, well below the average for Sub-Saharan Africa (0.38 in 1996). Nevertheless, in December 1999 the two top banks alone held about 42 percent of deposits and advances of the banking system. The top five accounted for 77 percent of credit and 72 percent of deposits. Public sector banks, which in 1991 accounted for about 70 percent of the credit and deposits of the banking system, have since lost their dominance but still account for 30 percent of credit and of deposits (Table 8.2).33
Indicators of the Size and Concentration of the Banking Sector
Banks with more than 60 percent private capital (Barclays Bank of Ghana, the local branch of Bank of Commerce and Credit International, and Continental Acceptances Ltd. are here considered as having mixed ownership).
Total credit to the private sector.
Indicators of the Size and Concentration of the Banking Sector
| Indicator | 1988 | 1991 | 1992 | 1998 | 1999 | |
|---|---|---|---|---|---|---|
| No. of banks in operation | 10 | 12 | 13 | 17 | 19 | |
| Privately owned1 | 2 | … | 4 | 9 | 11 | |
| No. of bank branches | 405 | 328 | … | 315 | … | |
| Per thousand inhabitants | 3.3 | 2.1 | … | 1.8 | … | |
| Percent owned by GCB | … | 84 | … | 43 | … | |
| Banking system assets (in percent of GDP) | 21 | 19 | 23 | 27 | 21 | |
| Nonperforming loans as share of total credit (in percent) | … | 41 | … | 17 | 13 | |
| Share of assets held by four largest banks (in percent) | 81 | 77 | 72 | 55 | … | |
| Public sector share (in percent) | ||||||
| Of total assets | 79 | 73 | 70 | 37 | … | |
| Of total capital | … | 77 | 76 | 34 | … | |
| Of total deposits | 73 | 71 | 65 | 32 | 30 | |
| Of total advances2 | 71 | 70 | 63 | 54 | 41 | |
| Of nonperforming loans | … | … | … | 84 | 77 | |
Banks with more than 60 percent private capital (Barclays Bank of Ghana, the local branch of Bank of Commerce and Credit International, and Continental Acceptances Ltd. are here considered as having mixed ownership).
Total credit to the private sector.
Indicators of the Size and Concentration of the Banking Sector
| Indicator | 1988 | 1991 | 1992 | 1998 | 1999 | |
|---|---|---|---|---|---|---|
| No. of banks in operation | 10 | 12 | 13 | 17 | 19 | |
| Privately owned1 | 2 | … | 4 | 9 | 11 | |
| No. of bank branches | 405 | 328 | … | 315 | … | |
| Per thousand inhabitants | 3.3 | 2.1 | … | 1.8 | … | |
| Percent owned by GCB | … | 84 | … | 43 | … | |
| Banking system assets (in percent of GDP) | 21 | 19 | 23 | 27 | 21 | |
| Nonperforming loans as share of total credit (in percent) | … | 41 | … | 17 | 13 | |
| Share of assets held by four largest banks (in percent) | 81 | 77 | 72 | 55 | … | |
| Public sector share (in percent) | ||||||
| Of total assets | 79 | 73 | 70 | 37 | … | |
| Of total capital | … | 77 | 76 | 34 | … | |
| Of total deposits | 73 | 71 | 65 | 32 | 30 | |
| Of total advances2 | 71 | 70 | 63 | 54 | 41 | |
| Of nonperforming loans | … | … | … | 84 | 77 | |
Banks with more than 60 percent private capital (Barclays Bank of Ghana, the local branch of Bank of Commerce and Credit International, and Continental Acceptances Ltd. are here considered as having mixed ownership).
Total credit to the private sector.
The average share of nonperforming loans in the banking system increased in the 1990s, to reach 17 percent of total credit to the private sector in 1998; about 78 percent of the total amount of these loans is held by the public sector banks. When one excludes from the calculation the troubled banks affected by the A-life fraud, the average drops considerably (to 12 percent). By the end of 1999 the share of nonperformling loans, excluding the troubled banks, had dropped to 11 percent. Most of the nonperforming loans continue to be held by state-owned banks (67 percent of the total, excluding the troubled banks) and represent close to 50 percent of total credit to the public extended by these banks. When one excludes public and mixed-ownership banks, the average drops to about 6 percent (Table 8.3). Most of the recent improvement in these ratios appears to depend on the rapid increase of private sector credit in the last two years. However, this increase has also been accompanied by an increase in the share of substandard loans. Loans that are past due represented 20 percent of total credit in December 1999 (Table 8.4), with an average share of 30 percent held by public sector banks and 17 percent by the rest of the system. It is hoped that this will not lead to a further accumulation of nonperforming loans in the near future.
Selected Banking System Indicators as of December 1998
(In percent except where noted otherwise)
Loans over 360 days overdue.
Loans over 30 days overdue.
As of January 1999.
As a percentage of bank net worth.
Selected Banking System Indicators as of December 1998
(In percent except where noted otherwise)
| Ownership Type | Nonperforming Loans as Share of Total Loans1 | Loans Past Due as Share of Total Loans2 | Net Open Foreign Exchange Position as Share of Total Loans3 | Capital Adequacy Ratios | Return on Assets4 | |
|---|---|---|---|---|---|---|
| State | 47.2 | 49.6 | 46.5 | -47.1 | -26.2 | |
| Private | 9.5 | 13.3 | 41.7 | -33.5 | 6.7 | |
| Mixed | 42.8 | 42.8 | 54.1 | -73.3 | 3.0 | |
| All banks | 17.2 | 18.8 | 47.4 | -28.9 | -11.6 | |
| Memorandum | ||||||
| Stock of nonperforming or past due loans (in billions of cedis) | 324.6 | 356.1 | ||||
Loans over 360 days overdue.
Loans over 30 days overdue.
As of January 1999.
As a percentage of bank net worth.
Selected Banking System Indicators as of December 1998
(In percent except where noted otherwise)
| Ownership Type | Nonperforming Loans as Share of Total Loans1 | Loans Past Due as Share of Total Loans2 | Net Open Foreign Exchange Position as Share of Total Loans3 | Capital Adequacy Ratios | Return on Assets4 | |
|---|---|---|---|---|---|---|
| State | 47.2 | 49.6 | 46.5 | -47.1 | -26.2 | |
| Private | 9.5 | 13.3 | 41.7 | -33.5 | 6.7 | |
| Mixed | 42.8 | 42.8 | 54.1 | -73.3 | 3.0 | |
| All banks | 17.2 | 18.8 | 47.4 | -28.9 | -11.6 | |
| Memorandum | ||||||
| Stock of nonperforming or past due loans (in billions of cedis) | 324.6 | 356.1 | ||||
Loans over 360 days overdue.
Loans over 30 days overdue.
As of January 1999.
As a percentage of bank net worth.
Selected Banking System Indicators as of December 19991
(In percent)
Data do not include BHC CooP Bank, and BCC, which were in liquidation.
Loans over 360 days overdue.
Loans over 30 days overdue.
Selected Banking System Indicators as of December 19991
(In percent)
| Ownership Type | Nonperforming Loans as Share of Total Loans2 | Loans Past Due as Share of Total Loans3 | Net Open Foreign Exchange Position as Share of Total Loans | Capital Adequacy Ratios |
|---|---|---|---|---|
| State | 22.2 | 31.1 | 44.5 | 14.8 |
| Private | 6.4 | 16.5 | 45.3 | 21.0 |
| Mixed | 12.6 | 14.5 | 28.5 | 9.3 |
| All banks | 13.0 | 20.3 | 40.3 | 11.5 |
| Stock of nonperforming or past due loans (in billions of cedis) | 386.7 | 604.4 |
Data do not include BHC CooP Bank, and BCC, which were in liquidation.
Loans over 360 days overdue.
Loans over 30 days overdue.
Selected Banking System Indicators as of December 19991
(In percent)
| Ownership Type | Nonperforming Loans as Share of Total Loans2 | Loans Past Due as Share of Total Loans3 | Net Open Foreign Exchange Position as Share of Total Loans | Capital Adequacy Ratios |
|---|---|---|---|---|
| State | 22.2 | 31.1 | 44.5 | 14.8 |
| Private | 6.4 | 16.5 | 45.3 | 21.0 |
| Mixed | 12.6 | 14.5 | 28.5 | 9.3 |
| All banks | 13.0 | 20.3 | 40.3 | 11.5 |
| Stock of nonperforming or past due loans (in billions of cedis) | 386.7 | 604.4 |
Data do not include BHC CooP Bank, and BCC, which were in liquidation.
Loans over 360 days overdue.
Loans over 30 days overdue.
Interest rates have scarcely responded to changes in the macroeconomic environment. In 1995 the average yearly inflation rate was about 50 percent, yet lending rates were about 40 percent, and savings rates barely reached 30 percent–both negative in real terms. In 1998 inflation more than halved, but lending rates remained stubbornly around 30–40 percent. The presence of state-owned banks with large shares of non-performing loans may be one reason for this behavior of interest rates. In general, demand deposits, which represent 33 percent of total deposits, yield little or no interest. This implies that, in the event of a downward adjustment in the interest rate received on credits, banks would not be able to adjust the interest rate paid on deposits. Thus their cost of funds would be mainly unaffected. This cost structure suggests that declines in interest rates are likely to reduce bank spreads, possibly to levels that would make the banks unprofitable. Finally, interest rates did start to decline in response to the lower inflation numbers in the last quarter of 1998, although real rates remained high throughout 1999. However, at the end of the year, rates were increased once again as a result of the rapid depreciation of the currency.
Another feature of the banking system in Ghana is its high dollarization: 34 percent of total assets and 29 percent of deposits were denominated in foreign currency as of December 1999. For borrowers the main attraction of foreign currency credits is the interest rate, which is much lower than that on domestic currency loans; in December 1999 interest rates on foreign currency loans were 27–34 percentage points lower than those on domestic currency loans. Lenders, meanwhile, have an incentive to lend in foreign currency because they need to maximize the rate of return on their foreign exchange deposits. Depositors still maintain a large share of their deposits in foreign currency both because of positive real interest rates (rates on foreign currency deposits range around 2–3 percent) and because of a continued lack of confidence in the sustainability of economic policies.
The dollarization of the banking system in Ghana has led the supervisory authorities to monitor closely the foreign exchange exposure of individual banks to ensure that they are not taking excessive risk. New foreign exchange exposure regulations were implemented in 1999, but the net open positions of many of the Ghanaian banks (calculated as the ratio of the difference between assets and liabilities denominated in foreign currency to shareholders’ funds) have remained high. The net open position of the consolidated banking system at the end of 1999 was about 40 percent of shareholders’ funds, an improvement of about 7 percentage points with respect to 1998, but still above the prudential requirement of 30 percent. In particular, public sector and merchant banks continued to have very high exposure ratios, above the regulatory limits. Therefore the banking system in Ghana is highly vulnerable to exchange rate movements, which could result in large losses in relation to shareholders funds.
How Far Have the Reforms Gone?
The current structure of the banking system suggests that some progress has been achieved since the first wave of reforms; nevertheless, much remains to be done. Banks seem to rely excessively on income from government securities, with loans to the private sector still representing a relatively small proportion of their assets. At an early stage in the banking reforms, the supervisory authorities eliminated non-performing loans from the consolidated balance sheet of the banking system at great cost.
However, in mid-1998 the average share of non-performing loans in total credits in the banking system was still unacceptably high at around 27 percent, only 14 percentage points below levels reached early in the first wave of banking reforms.34 Encouragingly, in the recent past this ratio has dropped further, to 11 percent in January 2000, after the closure of two ailing banks. There is reason to hope that the recent state enterprise divestitures and bank closures will prevent the recurrence of large accumulations of nonperforming loans. In addition, the bulk of non-performing assets currently in the system does not relate to credit to state enterprises, a widespread problem in the first wave of reforms.
However, the relatively rapid accumulation and large share of loans that have become past due in the context of rapidly expanding credit to the private sector are also worrisome. One small, newly established bank already counted 37 percent of its loans as past due. The two remaining state-owned banks had an average share of 46 percent of their loans past due.
The fact that state-owned banks continue to have a significantly higher share of substandard loans than did privately owned banks in the past has slowed the pace of the divestiture program and may have endangered the success of the reforms. This may reduce significantly, if not eliminate, the proceeds realized by the government from the divestiture.
On the positive side, the restructuring program has tended to increase the average capitalization of banks; in particular, the swapping of bad loans and the enforcement of new regulations in the early 1990s succeeded in improving capital adequacy ratios in the system. Except for three banks that were in liquidation as of January 2000, all other banks meet or exceed the minimum capital requirement, which is 6 percent of assets.35
The regulatory framework developed in 1989 is for the most part in line with the core principles established by the Basle Committee on Banking Supervision. However, it needs to be implemented with greater consistency and vigor. Penalties for noncompliance were stated in nominal terms and are now negligible after years of high inflation. In addition, the supervisory authorities have not enforced bank regulations as strictly as they could have: penalties for noncompliance have at times been suspended on an ad hoc basis. For example, several banks presently do not comply with single-customer exposure limits and with the limit on foreign exchange exposure. Moreover, improvements in the payments system have been slow, leaving the system open to fraud, as the 1997 A-life incident demonstrated (see Box 8.1). However, the authorities have recently renewed their efforts to modernize the payments system by standardizing checks and using magnetic ink to permit the automation of most clearing procedures.
Finally, the revision of the legal framework has not significantly increased the speed with which commercial banks can use the judicial system to recover losses associated with bad credit. This remains an important obstacle to the development of the banking business.
Lessons Learned and Next Steps
Ghana’s early banking reforms were considered among the most successful in Africa (World Bank, 1994). Yet the banking system is still not as healthy and competitive as one would expect six years after these reforms were completed. The main problem has been that the divestiture program, which was seen as the second phase of the reforms, has progressed very slowly. Therefore most state-owned banks returned to the same practices that had been at the root of the banking system’s problems in the 1980s: excessive political influence, weak management, backward information and accounting systems, inadequate credit approval procedures, and poor internal controls. As a result, nonperforming loans increased once again, further slowing the divestiture process by reducing the attractiveness of these banks to potential investors.
With the closure of the three banks that have failed to meet capital adequacy requirements, the government will have made a substantial step forward in its reform of the financial sector. However, to prevent the perpetuation of some of the existing problems and the recurrence of the old ones, the divestiture process needs to be completed. This means the privatization of all state-owned banks and the complete divestiture of the shares of both the government and the central bank in all commercial banks. The government should also ensure that the shares held by the social security trust fund remain a minority (passive) investment.
Second, competition should be encouraged by allowing new, professionally managed banks to enter the market. However, only banks with proven, well-qualified, and professional management should be allowed to enter. The supervisory authorities should also increase the monitoring of loan quality and insist on strict adherence to procedures to ensure the soundness of the rapidly expanding system.
Third, prudential regulations should be updated and their enforcement strengthened. In particular, foreign exchange exposure limits should be enforced strictly, and the new banking law should be implemented as soon as it receives parliamentary approval. Regulations on single-creditor risk also need to be enforced. On-site and off-site supervision needs to be reinforced through close monitoring of banking procedures. The use of external auditors may also be considered. Penalties for noncompliance with prudential regulations need to be made more meaningful and fully enforced.
Finally, the government needs to improve the environment for banking business. In addition to ensuring a stable macroeconomic environment, the government will need to strengthen the legal framework and the judicial system to facilitate the recovery of nonperforming loans, while protecting borrowers from unfair practices. The payments system will also need to be modernized to prevent fraud and facilitate financial transactions.
