United Republic of Tanzania
Financial System Stability Assessment Update
Author: International Monetary Fund
Publication Date:
23 Jun 2010
eISBN:
9781455204199
Language:
English
Keywords:
ISCR; CR; bank; loan; capital market; market; Tanzania; government bond; interest rate
Download PDF (2.8 MB)

The global recession has had a significant impact on economic performance and the outlook in Tanzania. Tanzanian banks were not directly affected by the global financial crisis, but second-round effects of the crisis have intensified the risks to the financial sector. Domestic liquidity tightened considerably at the onset of the global financial crisis but has since eased, owing to a significant loosening of monetary policy to help mitigate the economic downturn. The policy response of the authorities included a rescue package designed to support sectors affected by the crisis.

Abstract

The global recession has had a significant impact on economic performance and the outlook in Tanzania. Tanzanian banks were not directly affected by the global financial crisis, but second-round effects of the crisis have intensified the risks to the financial sector. Domestic liquidity tightened considerably at the onset of the global financial crisis but has since eased, owing to a significant loosening of monetary policy to help mitigate the economic downturn. The policy response of the authorities included a rescue package designed to support sectors affected by the crisis.

I. Macroeconomic Environment and Financial Structure

A. Macroeconomic Environment

8. The global recession has had a significant impact on economic performance and the outlook in Tanzania (Table 1). Export sectors led a decline in real GDP growth to about 5.5 percent in 2009, from some 7½ percent in 2008, while key sources of capital—notably FDI and syndicated loans from abroad—have contracted, compounding the effects on the balance of payments of portfolio outflows. Better performance in some sectors (including construction, telecommunications, and food crops) and a fiscal stimulus of 5 percent of GDP is helping to moderate the slowdown with GDP growth expected to pick up in 2010, but there are still downside risks to the outlook.

Table 1.

Tanzania: Selected Economic and Financial Indicators

(2006/07–2011/12)

article image
Sources: Tanzanian authorities; and Fund staff estimates and projections.

Data are on calendar year basis. For example, 2006/07 data are for calendar year 2006.

Figures for 2008/09 onward reflect a new series based on the Fund’s 2000 Monetary and Financial Statistics Manual.

End-year (June) monthly weighted-average yield of 35-, 91-, 182-, and 364-day treasury bills. For 2009/10, figure is for February 2010.

Including change in stocks.

From the sixth review under the PSI, based on previous GDP projection.

9. Tanzanian banks were not directly affected by the global financial crisis, but second-round effects of the crisis have intensified the risks to the financial sector (see Risk Assessment Matrix in Appendix II). Going into the crisis, banks were well-capitalized and not exposed to the asset classes that led to large losses in major financial centers. However, exposure to distressed sectors (including cash crops, tourism, and transportation) has affected loan quality in a number of banks—including some larger institutions. Other financial institutions, such as the rapidly growing and inadequately supervised pension funds, could also be affected by a deterioration of asset quality, including in the real estate sector, which make up a growing part of asset portfolios.

10. Domestic liquidity tightened considerably at the onset of the global financial crisis but has since eased, owing to a significant loosening of monetary policy to help mitigate the economic downturn. In January 2009 the BOT increased the reserve requirement on government deposits held in commercial banks and no longer permitted the use of cash-in-vault to meet required reserves, leading to liquidity shortages for some banks and a spike in money market rates. Liquidity returned to ample levels in April 2009, when the BOT eased its monetary policy stance to encourage the flow of credit.

11. The policy response of the authorities included a rescue package designed to support sectors affected by the crisis.3 The mission noted the risks related to market distortion and moral hazard of compensating banks for loan losses; in response, financial support was offered to cotton and coffee traders directly, rather than to the banks. Funds were nevertheless channeled through the banks, however, to ensure that they were used for resolving nonperforming loans (NPLs) rather than for other purposes.

B. Financial Structure

12. The banking system in Tanzania has grown significantly since 2003, but remains relatively small and dominated by a top tier of larger domestic legacy and foreign banks (Table 2). The top tier mainly caters to a small group of large corporates—which often represent up to 70 percent of banks’ loan portfolios—leaving the retail market underserved. Government ownership is limited to four smaller fully-owned banks and minority stakes in the three largest domestic banks. Tanzanian banks have very limited operations abroad (one bank is active in Cyprus), and linkages with other financial institutions are largely absent.

Table 2.

Tanzania: Financial System Structure, 2006–09

article image
Source: BOT.

NBC, NMB and CRDB

Citibank Standard Chartered, Stanbic, and Barclays

Rural and cooperative banks, as defined under the banking and Financial Institutions Act, 2006

PPF, LAPF, GEPF, ZSSF

Money market mutual funds

Source: BOT.

13. Overall efficiency in the banking system remains weak, but especially the larger banks are profitable as a result of high interest margins. Since 2003, the banking system’s overhead costs to total assets ratio has not significantly declined and remains high at 5.7 percent (although this is in line with the regional average). Large banks’ networks allow them to mobilize ample, low-cost deposits at interest rates of between 1 percent and 2.5 percent, while a significant portion of assets is still invested in government securities with yields up to 20 percent (Figure 1), leading to high interest margins (Figure 2). As a result, the three largest banks achieved a return on assets in 2008 of 3.1 percent, exceeding the regional average of 2.5 percent.

Figure 1.
Figure 1.

Tanzania: Banking System Asset Composition, 2001–09

(In percent)

Citation: IMF Staff Country Reports 2010, 177; 10.5089/9781455204199.002.A001

Source: BOT.
Figure 2.
Figure 2.

Tanzania: Net Interest Margin, 2008

(In percent)

Citation: IMF Staff Country Reports 2010, 177; 10.5089/9781455204199.002.A001

Source: Bankscope.

14. Smaller banks, lacking extended branch networks, have been unable to compete with the larger banks and concentration remains high. Since 2003, the asset share of the three largest banks (all previously government-owned) only declined slightly, from 66 percent to 64 percent. The BOT could promote consolidation of medium-sized banks to contest the top tier by judiciously raising minimum capital requirements. In addition, the current 80 percent loan to deposit ceiling puts the smaller banks at a disadvantage and the authorities could consider removing it. It is an unusual prudential requirement, the requirement is frequently waived, and 8 banks exceeded it as of end-2009.

C. Financial Depth and Access4

15. Despite rapid asset growth in recent years, private credit and deposit mobilization levels still lag behind countries in the region (Figure 3). The ratio of private credit to GDP rose from 5 to 16 percent of GDP during 2003–09, but remains below the regional average of 28 percent. Similarly, domestic deposits increased from 15 to 25 percent of GDP in the same period, compared to the regional average of 44 percent. Factors that have impeded a further increase in intermediation include a poor business environment, historically high interest rates on government securities, low competition, high profitability, lack of longer-term funding, and the loan-to-deposit ceiling.

Figure 3.
Figure 3.

Tanzania: Private Credit and Domestic Deposits as a Share of GDP, 2008

Citation: IMF Staff Country Reports 2010, 177; 10.5089/9781455204199.002.A001

Source: BOT.

16. Only one in six Tanzanians have access to financial services from formal institutions, which compares poorly to the country’s peers (Figure 4). This share has increased since 2006, however, when formal access stood at only 11 percent. An additional 27 percent of the population is served by informal village associations, but over half of the population is financially excluded; this share has remained largely unchanged since 2006. Rural areas are particularly underserved: the rural parts of most regions reported an access rate of less than 10 percent.

Figure 4.
Figure 4.

Tanzania: Access to Financial Services Across Sub-Saharan Africa

(In percent)

Citation: IMF Staff Country Reports 2010, 177; 10.5089/9781455204199.002.A001

Source: FinScope.

17. The lack of a reliable credit information system and weak creditor rights restrict credit access and contribute to a high cost of credit. The procedures to create security interests are less than reliable and cumbersome, and the enforcement of claims is lengthy, unpredictable, and suffers from legal complications that can be used by debtors to delay proceedings. Credit information and creditor rights could be substantively enhanced with a series of legal and institutional reforms that include: (i) the creation of a Credit Bureau with mechanisms to uniquely identify debtors; (ii) granting exclusive jurisdiction over enforcement of creditor claims to commercial courts; and (iii) implementing a comprehensive reform of civil procedure to reduce use of delay tactics by debtors, including by limiting the number of available appeals.

II. Safeguarding Financial Stability

A. Risks and Vulnerabilities in the Banking System

18. The 2003 FSAP concluded that banks were generally liquid, well-capitalized, and resilient to most shocks. The system had a capital ratio in excess of 20 percent, and relatively low lending activity limited the extent of credit risk, while exchange-rate risk was well-contained. The main vulnerability stemmed from interest-rate risk, with banks holding a significant share of assets in government bonds.

19. The FSAP update mission found that data deficiencies hampered its analysis of risks in the financial sector and the authorities are urged to improve data quality and timeliness. Data provided to the mission on NPLs and provisions contained inconsistencies, including individual bank data that appeared not to correspond with available information about prudential compliance. The mission was unable to obtain detailed data on smaller banks, where weaknesses are concentrated. It is recommended that the authorities compile a consistent and timely data set for macroprudential analysis, including stress tests. The new system of electronic transmission of prudential data by banks could provide a sound basis for such a set, but data verification, retrieval and analysis needs to be improved. If necessary, the authorities are urged to seek further technical assistance in this area.

20. While available financial soundness indicators (FSIs) (Table 3) show a banking sector that is on average still adequately capitalized, weaknesses exist in some (mostly smaller) banks. The system-average Tier I and Tier II Capital Adequacy Ratios (CARs) rose by almost 4 percentage points between 2008 and 2009, to between 18 percent and 19 percent for both (compared to prudential minima of 10 percent (Tier I) and 12 percent (Tier II)).5 The average CAR masks weakness in some systemically less important banks, with three recently falling below the required minimum.

Table 3.

Tanzania: Selected Financial Soundness Indicators,

2006–09

article image
Source: BOT.

21. Vulnerabilities are heightened in part by uneven compliance with prudential requirements (Table 4), and the BoT should be more proactive in ensuring compliance. The three undercapitalized banks at end-June were required to file a recapitalization plan with the BOT with a time-bound path to regain compliance. Although such corrective action is appropriate, it is recommended that the BOT take a more proactive stance to ensure that incipient problems in banks are addressed before they cause a breach of prudential limits. This would be especially important in the case of deteriorating CARs.

Table 4.

Tanzania: Compliance with Prudential Limits, end-December 2009

article image
Source: BOT.

22. Underprovisioning by banks in particular appears widespread. While a lack of data prevents a full analysis, a substantial minority of the banks appear not fully to comply with the prudential provisioning requirements (Table 5). A simulation to bring provisions up to a level believed to be closer to prudential compliance (lack of data on the impact of government guarantees and collateral prevent an exact calculation) shows that six banks would be undercapitalized if they were to provision adequately, although most of these are not of systemic importance and total recapitalization costs would be small (totaling T Sh. 5 billion—0.017 percent of GDP—for the six banks combined). Supervisors should work with banks to improve the quality of provisioning data and ensure compliance with the prudential requirements.

Table 5.

Tanzania: Adequacy of Loan Loss Provisions, December 2009

article image
Source: BOT and IMF staff estimates.

23. Risk in banks is focused around credit risk after rapid credit growth in recent years. Between 2003 and 2008 strong growth in private sector credit (Figure 5) raised the share of loans in banks’ portfolios from 32 percent to 54 percent, exposing the sector to increased credit risk. In addition, the high rates of credit growth could signal a trend towards less robust loan underwriting practices and raises concerns about the possibility of a future deterioration in loan portfolio quality.

Figure 5.
Figure 5.

Tanzania. Selected Financial Sector and Market Developments 2002–09

Citation: IMF Staff Country Reports 2010, 177; 10.5089/9781455204199.002.A001

Source: BOT.

24. Reported NPL rates peaked in mid-2009, but they exhibited large variation among banks, partly reflecting different exposures to distressed sectors. System-wide NPLs rose to 7.8 percent in June 2009 from 6.2 percent in December 2008, but subsequently declined to 6.7 percent, partly as a result of the government rescue package for distressed borrowers. NPL rates vary greatly across banks: for the systemically important banks at end-June 2009, they ranged from 1.8 percent to 14.2 percent, with two banks reporting rates in excess of 10 percent. Much of the variation reflects different sectoral distributions of loans, especially in the agricultural and transportation sectors.

25. Stress tests indeed confirm that the main vulnerability for banks stems from credit risk, particularly from large exposures. The stress tests found the following:6

  • Substantial concentration in the loan portfolio leaves banks exposed to the failure of large borrowers (Table 6). Failure of the single largest exposure renders five of the largest 10 banks undercapitalized (with one insolvent), with a drop in the system-wide CAR of more than one-third to below 12 percent. The undercapitalized banks would require additional capital of 0.2 percent of GDP to return to the regulatory minimum capital level. It was reported, however, that the exposure of banks was limited through government guarantees or cash collateral covering some of these loans. The mission could not verify the extent to which this mitigated the impact of potential loan failures on banks, as it was unable to obtain accurate data on guarantees and collateral.

  • A sharp increase in short-term rates from today’s very low levels would hurt a number of banks (Table 7). Were the yield curve to flatten out to where both short and long rates stand at 20 percent (reflecting increases of some 6 percentage points in long rates and 18 percentage points in short rates), a number of larger banks would show a sharp drop in net interest income, as mostly longer maturity assets would see yields rise less (and later) than mostly shorter maturity liabilities.7

  • Foreign exchange risk is contained by relatively prudent management of open positions by banks.8 In stress tests, most banks would easily withstand appreciations and depreciations of 30 percent, partly as a result of prudent management of open positions. In addition, indirect (credit) risk is contained because borrowers in foreign currencies are concentrated in the tradable sectors, with exposures hedged through income in foreign currency.

  • Liquidity stress tests confirm the ample liquidity in the system (Table 8). All banks are able to withstand a liquidity drain of 34 percent for demand deposits and 16 percent for time deposits in one day without recourse to interbank lending or emergency liquidity from the BOT. Most larger banks, except one medium-sized foreign subsidiary, could withstand a second day of such a substantial drain. Banks would cope well with a withdrawal of non-resident deposits, as these make up only a small proportion of total deposits.

  • The comparison across all stress tests reveals that there are a number of consistently weaker (mostly smaller) banks, which also suffer in more comprehensive multifactor shocks. Banks with lower initial capital buffers suffer in most stress tests—enough to become undercapitalized. One larger foreign bank also suffers in two simulations of adverse macroeconomic scenarios, originating in (i) a drop in commodity prices and (ii) a sudden capital outflow.

Table 6.

Tanzania: Stress Test Results—Default of Three Largest Borrowers

article image
Source: BOT.
Table 7.

Tanzania: Summary of Stress Test Results

article image
Source: IMF staff estimates.Notes:

Banks 1–5: large foreign banks; banks 6–8: large domestic banks; banks 9–25: medium and small-sized commercial banks; banks 26–34: community banks.

Table 8.

Tanzania: Liquidity Stress Test Results

article image
Source: IMF Staff estimates.

26. The stress test exercise in particular suffered from the previously mentioned data limitations. Detailed data on the loan portfolio for some smaller banks was not available, forcing exclusion of the smaller banks from some stress tests. Individual bank data that was available suffered from inconsistencies and unexplained outliers affecting the results. The data weaknesses suggest a considerable need for caution in interpreting the stress testing results, which should be used primarily as a basis for further prudential investigation of the revealed vulnerabilities.

B. Crisis Management Framework

27. Important crisis management tools and components of a financial safety net are in place, but gaps remain and a comprehensive framework for crisis management is lacking. While there is a deposit insurance scheme and a regulatory framework for banking resolution, arrangements for emergency liquidity provision could be improved. Other key enhancements to preparedness for crisis management would include formal contingency plans, explicit institutional arrangements for cooperation among government agencies involved in crisis management, and regular crisis simulation exercises. Steps are being taken by the authorities to address these issues with technical assistance from the Fund.

28. The operational framework for the BOT’s role as a liquidity provider to banks is adequate, but a framework for Emergency Liquidity Assistance (ELA) is yet to be developed. In day-to day operations, the BOT offers two lending facilities: (i) an intraday credit facility for payments system purposes and (ii) an overnight (Lombard) credit facility carrying a penalty interest rate (see Box 2 in section D). In both cases, borrowing is accessible for all commercial banks against pledged collateral in the form of short-term government securities. There are no explicit operational arrangements, however, for ELA beyond the day-to-day liquidity management framework and tools. The BOT should develop an operational framework for: (i) ELA in case of an acute liquidity problem in normal times (assistance to be given only to illiquid but solvent banks against adequate collateral); (ii) ELA in a systemic crisis; and (iii) assistance to a critically undercapitalized but systemically important bank—including arrangements with the Ministry of Finance and Economic Affairs (MOFEA) in case of potential fiscal implications. The BOT should communicate these principles to the banks.

29. The framework for bank intervention and resolution appears adequate for a single bank failure, but a systemic crisis may require further arrangements. The secondary legislation that took effect in early 2009 provides the BOT with adequate authority to undertake enforcement action when a bank is facing financial difficulties or is operating in an unsound manner. It may give general or specific directions to a bank to affect its operations or the national interest, and the bank is required to comply. Possible prompt corrective actions in case of financial difficulties include monetary penalties, removal of personnel, appointment of a statutory manager, and license revocation. Private solutions (such as injection of capital by new or existing shareholders) are also considered. The framework for dealing with systemic crises may be enhanced, however, by making explicit logistical arrangements for communication and cooperation with the MOFEA and other supervisory agencies during a crisis, and allowing for more discretionary action to contain the crisis.

30. The new legal framework for deposit insurance adopted in 2006 is a welcome improvement, but additional measures would further enhance the safety net. In line with the 2003 FSAP recommendations, the Deposit Insurance Fund (DIF) has largely been confined to a pay-box (i.e., its role is restricted to repaying depositors in case of a bank failure), but the Deposit Insurance Board retains extended functions, such as liquidation responsibilities and participation in onsite examinations. Deposits (in domestic and foreign currency) are insured up to the equivalent of about US$350, but interbank, government, and investment companies’ deposits are excluded. The maximum coverage is low compared to neighboring countries, risking a deposit drain abroad when Tanzania liberalizes the capital account (see section III.B). While the recent initiative of the Board to triple coverage to the equivalent of about US$1,000 is welcome, reserves are relatively small (less than 1 percent of banks’ deposit liabilities) and the failure of a systemically important bank would overwhelm resources. It is recommended that reserves be strengthened (in part with the aid of the recent contribution increase) to enhance the DIF’s ability to deal with the failure of a larger bank; a contingency plan with funding options (e.g., loans, public funds, or ex-post bank levy) for such an eventuality could also be helpful.

31. While important measures have been taken by the BOT to strengthen the day-to-day monitoring of global and domestic developments, it is recommended that a formal crisis management plan be adopted. The BOT has daily market monitoring meetings with senior management and a Financial Sector Stability Department has been established to coordinate work on financial sector stability and crisis management, including production of Financial Stability Reports. Beyond these measures, a formal crisis management plan would enhance crisis preparedness, and a crisis simulation exercise could identify weaknesses in present crisis management tools, in the coordination with other government agencies, and in the crisis communication strategy. The possibility of participating in cross-border contingency planning exercises should be explored.

C. Banking Sector Regulation and Supervision

32. The BCP Assessment undertaken in the context of the 2003 FSAP found that the framework of banking supervision was generally adequate, but that supervision needed to be put on a more risk-based footing. It was also recommended that transparency of banking supervision policies be enhanced and that the BOT be given independence from the government with respect to revocation of licenses and instituting corrective action. The crisis management framework and financial safety net needed strengthening.

33. Substantial progress has been made in aligning the legislative framework underpinning banking supervision to the BCP. It provides the BOT sufficient powers to discharge its mandate as the authority responsible for licensing, supervising and regulating the banking industry. Licensing is now the sole preserve of the BOT, removing the potential for political interference and review of licensing files indicates a detailed and thorough approach by the BOT. More recently, legislation has been supported by a suite of prudential regulations and guidelines. To improve offsite surveillance, all bank reporting to the BOT has been automated and is captured in the Banking Supervision Information System (BSIS).

34. In contrast, progress in enhancing operational aspects of supervision has been hampered by important capacity and organizational challenges. The Banking Supervision Department is operating with only half the authorized number of staff, while scarce resources continue to be consumed by activities marginal to supervision. Skill levels and additional skills absorption also present challenges, complicating the successive introduction of a CAMELS rating system (2005) and a risk-based supervision approach (2007). Important deficiencies exist in work processes, including lengthy clearance procedures for communication with banks and a failure to organize and safely store working papers completed in the course of inspections.

35. While substantial efforts have been made, the issues raised above leave BCPs compliance gaps (Appendix IV), despite a positive self-assessment carried out by the BOT. To promote full compliance with the BCPs, it is recommended that management consider the following measures:

  • Address capacity and organizational challenges. In addition to training of existing staff, recruitment efforts should be stepped up as much as possible to fill authorized capacity with qualified staff with minimum required supervisory and financial analysis skills. Other organizational improvements, including reducing clearance requirements for communication with banks, are key. Work pressures should not be allowed to cause deterioration in orderly supervisory processes, including proper documentation, filing, and storage.

  • Obtain hands-on technical assistance in the conduct of risk-based supervision to achieve full effectiveness of the BOT’s risk-based framework.

  • Increase cooperation with foreign regulators. Supervisory cooperation with EAC partner agencies appears strong and links are developing with another jurisdiction where Tanzanian banks operate. It is recommended that the BOT intensify cooperation with regulators of foreign bank subsidiaries operating in Tanzania and obtain an annual letter from “home” regulators confirming the parent bank’s financial condition.

  • Identify where correspondence between the BOT and banks may be entrusted to bank supervisors, without the need to obtain senior executive approval.

  • Engage banks’ senior management and their external auditors. In particular:

    • Require that senior management of a bank meet the BOT at least annually to present a full review of performance. Banks could also be required to submit a commentary on significant changes in prudential reporting figures at the time of submission. Regular tri-partite discussions with the bank’s management and its auditors may also prove beneficial.

    • Require that banks’ external auditors provide certification that prudential reports have been prepared correctly and reflect the institution’s records. This would eliminate the current practice whereby onsite examinations incorporate checking of prudential reports against the bank’s general ledger.

  • Eliminate the requirement that the opening of a new bank branch involves a physical inspection of (and report on) the premises by the BOT. This practice appears to add little value, but consumes scarce resources. The BOT has sufficient corrective powers if a bank is found, in the course of regular onsite inspections, to be conducting business in an unsafe manner.

D. Systemic Liquidity Management

36. The primary task of liquidity management in Tanzania is the sterilization of large liquidity injections resulting from government spending financed by direct budget support from donors. On average, Tanzania receives about 10 percent of GDP (or nearly US$2 billion) a year in donor assistance, more than half of which is provided as direct budget support that is deposited directly into the government’s account at the BOT (in Tanzanian shillings), while the FX counterpart is added to official international reserves. As the government expends these deposits, the resulting liquidity injection is mopped up by the BOT (Box 2).

37. In line with the 2003 FSAP recommendations, the use of liquidity management instruments was clarified to foster consistency in policy implementation. Operational changes in October 2007 led to a more balanced sterilization framework, with the BOT entering the FX market mostly on the sell side. Under the new policy, after calculating the sterilization needs for the month ahead, the BOT sells a proportional amount of FX on a daily basis. Additional interventions in the exchange market are limited to operations to smooth exchange rate movements (Figure 6).9

Figure 6.
Figure 6.

Tanzania: The Policy Mix, January 2006–December 2009

Citation: IMF Staff Country Reports 2010, 177; 10.5089/9781455204199.002.A001

Source: BOT.

Monetary Policy Tools of the Bank of Tanzania

The BOT has appropriate instruments available for withdrawing liquidity:

  • SMRs—10 percent on nongovernment deposits and 20 percent on government deposits (from January 2009), unremunerated, with averaging allowed over the previous 2-week fulfillment period. For banks with a sufficiently extensive branch network, the SMRs are reduced by 20 percent (following the exclusion from January 2009 of cash-in-vault for the SMRs calculation). The SMRs on FX deposits must be met in Tanzanian shillings.

  • Liquidity paper (LP). Treasury bills (t-bills) of 35, 91, 182, and 364 days; auctions are held fortnightly. The BOT sets the auction sizes for LPs mainly to roll over outstanding LPs over time, in line with limited government domestic financing requirements, and to mop up liquidity as necessary in excess of FX sales. The interest cost of t-bills issued for monetary policy purposes is shared by the BOT, which reimburses the government for this cost, up to a yearly maximum of 15 percent of the BOT’s gross earnings.

  • Sales of FX by the BOT in the interbank FX market.

  • REPOs (7- and 14-day maturities). REPO transactions are conducted through an auction process and take place on a daily basis to manage systemic liquidity between LP auctions.

  • Standing facilities. The BOT offers a Lombard facility, with a rate based on 35-day t-bills, overnight and repo rates, whichever is higher, plus a 20 percent penalty. There is a limited set of eligible collateral, with a maximum of 91 days to maturity. In addition, t-bills and bonds can be rediscounted at a 500 basis point penalty rate. A deposit facility is also periodically offered by the BOT.

38. The institutional framework for monetary policy decision-making appears appropriate, and communication to the market has recently improved. The governor’s monetary policy statement to parliament prior to the beginning of the fiscal year and the midyear policy review set out the broad monetary objectives for the year and the medium term. These are complemented by operational objectives set at monthly meetings of the Monetary Policy Committee. Weekly meetings of the Liquidity Management Committee, chaired by the governor, then set daily objectives for the week ahead, which are monitored and sometimes adjusted on a daily basis on a technical level, with approval by management. The recent initiative of the governor to meet monthly with chief executive officers (CEOs) of major banks about monetary policy implementation is a welcome improvement to the communication strategy. These technical meetings might be supplemented with less frequent high-level meetings about other strategic issues of interest to the banks.

39. While the operational changes of October 2007 were designed to foster more stable conditions in the money market (Figure 7), implementation could be further improved to increase the transparency of policy intentions. Recommendations are the following:

Figure 7.
Figure 7.

Tanzania: Money Market Transactions, January 2006–December 2009

Citation: IMF Staff Country Reports 2010, 177; 10.5089/9781455204199.002.A001

Source: BOT.

  • Distinguish clearly between sterilization and intervention objectives in procedures for selling FX. Sterilization sales should be predictable and transparent to avoid undue exchange-rate effects, and it is recommended they be moved to an auction format. Intervention transactions should be designed to have maximum impact on the exchange rate; continuing the use of the interbank market is therefore appropriate.

  • Use the SMR only for long-term structural sterilization of liquidity, with frequent changes to be avoided. The recent move to allow averaging within the reserve fulfillment period is welcome.

  • Use REPOs only for fine-tuning operations. Prolonged use of (reverse) REPOs can cause confusion about monetary policy intentions, suggesting that the often sizeable mopping up achieved is intended to be only temporary. More permanent instruments, such as FX sales or additional issuances of LP would be more appropriate, even if there is some impact on the shilling exchange rate or interest rates—indeed, such effects may be necessary to achieve the desired macroeconomic effects to underpin the sterilization effects.

  • Improve further the consistency between the announced t-bill offering and actual sales and refrain from discretionary behavior at auctions. Currently, bids for t-bills may be rejected even though the full offering has not been met, while bids for some maturities are accepted beyond the volume offered. Occasionally, the BOT approaches banks after the auction to conduct additional sales of t-bills. This tends to reduce the clarity of policy intentions.

40. In addition, it is recommended that the authorities consider measures to improve further the accuracy of liquidity forecasts, which are the lynchpin of effective liquidity management:

  • As government is the primary source of liquidity injections, advanced coordination between the MOFEA and BOT on planned major expenditures (as well as expected revenues) would help reduce forecasting errors. Specifically, cyclical patterns of expenditures (like monthly wage payments) and revenues (like end-month and end-quarter tax payments) could be taken into account more effectively.

  • Continued research into (excess) liquidity demand and monetary transmission mechanisms would enable the BOT to better align the supply of liquidity to demand patterns, helping to minimize the perceived need for banks to hold excess reserves and enhancing the functioning of the money market.

  • Further analysis of forecasting errors may identify other sources of liquidity demand and supply that should be taken into account in the forecasting exercise.

E. The Payments System

41. Substantial progress has been made towards improving the payment systems infrastructure since 2003. With the implementation of the Tanzania Interbank Settlement System (TISS) in 2004, large-value and time-critical payments are now handled in real time. The Bank of Tanzania Act of 2006 gives the BOT broad powers to regulate the TISS and other payment systems. By 2010/11, the BOT plans to implement a national interbank switch that will promote efficiency and interoperability for point of sale (POS) and electronic funds transfers (EFT) transactions. It is recommended that the framework for the national switch at a minimum include (i) mandatory participation for banks that issue payment cards; (ii) regulation of the fee structure for automated teller machines (ATMs) and EFT/POS transactions; and, (iii) no restrictions for participation in other payment card systems.

42. Addressing remaining gaps in the payments system legislation is critical to enable the TISS participants to understand and manage credit and liquidity risks. The BOT currently has limited legal powers to develop regulations and enforce sanctions with respect to the TISS. The authorities should enact a National Payment Systems Act to provide a clear legal definition of “irrevocability of orders” and “finality” of payments in the TISS, and to make BOT regulations in this area legally enforceable.

III. Longer-Term Finance

A. Capital Markets

43. The Tanzanian capital market is still at a nascent stage and has not played a major role in resource mobilization and long-term financing of the economy. Only fifteen companies are listed on the DSE, with a market capitalization of US$3.75 billion in February 2010 and a market turnover of US$9.6 million in the fourth quarter of 2009. The corporate bond market is very limited; as of February 2010 there were seven corporate bonds with an outstanding amount of US$58.7 million. On the government bond market, secondary transactions are limited, but recent oversubscriptions on the 10-year primary bond auctions and increased placements indicate a growing appetite for long-term instruments by investors (Figure 8).

Figure 8.
Figure 8.

Tanzania: Yield Curve for Government Bonds and Volume of 10-Year Bonds

Citation: IMF Staff Country Reports 2010, 177; 10.5089/9781455204199.002.A001

Source: BOT.

44. The regionalization of the EAC capital market presents considerable opportunities for Tanzania. An EAC regional capital market (currently planned for implementation by 2012) will enable investors to diversify their portfolios, better manage risks, and grant issuers access to a deeper pool of long-term resources to finance their investments. It will generate economies of scale in market infrastructures and increased efficiency in securities transactions through the interconnection of exchanges, Central Securities Depositories (CSDs) and payment systems across borders in the EAC region.

45. Development of the securities markets (including through regionalization) also presents significant challenges. It is recommended that the authorities:

  • Improve risk management for the settlement system. Settlement risk is currently significant and inhibits market development. Measures should include requiring guarantees by settlement banks of brokers’ obligations; required initial and variation margin payments from settlement banks and the introduction of a guarantee fund.

  • Strengthen the market legal and regulatory framework and pave the way for regional harmonization. Key priorities are to (i) adopt legislation requiring the demutualization of the DSE and the splitting of the DSE and the CSD into separate corporations; (ii) remove limitations on foreign investment in listed corporations; (iii) remove restrictions on foreign participation in the domestic government and corporate bond market; (iv) regulate over-the-counter (OTC) trade reporting and to allow OTC trading for bonds; (v) adopt regulations implementing common licensing standards for market intermediaries in the framework of the East African Member States Securities and Regulatory Authorities (EASRA); (vi) adopt regulations implementing common prudential standards for institutional investors, to be agreed in the framework of the EASRA; (vii) pass double-tax treaties with other partner states; and (viii) proceed with capital account liberalization (see below).

  • Overhaul the Capital Markets and Securities Act to strengthen independence and confer adequate inspection and investigation powers on the Capital Markets and Securities Authority (CMSA) and to allow international information sharing. The CMSA should add to its staff total and should focus on meeting the IOSCO Principles. It should further enhance collaboration with counterpart regulatory authorities in other EAC partner states, including by (i) establishing a single certification for brokers operating in the EAC; and (ii) drafting a memorandum of understanding (MOU) with EAC’s partner regulators that defines home-host supervisory responsibilities for capital market intermediaries, exchanges, CSDs, and institutional investors.

B. Capital Account Liberalization

46. Capital account liberalization is key to the development of the regional capital market. As of now the Tanzanian capital account is substantially controlled. Residents’ investments abroad and most inward capital transactions (except for FDI and some foreign loans) require the prior approval of the BOT and there is a general repatriation requirement for foreign currency receipts. Limits apply to the purchase of shares and corporate bonds on the DSE by nonresidents. The external use of the Tanzanian shilling is prohibited in most cases. These capital controls may have helped to mitigate the effect of the financial crisis on Tanzania, but they also introduce distortions, including a relatively low level of foreign investment in Tanzania and interference with the efficient allocation of resources and the development of financial markets.

47. In the context of the regional integration initiative, the BOT has developed a plan for the gradual lifting of capital controls by 2015. While the overall planned timeframe appears feasible, it is recommended that the liberalization plan be reviewed to ensure that the lifting of the controls is properly sequenced and coordinated with other supporting policies. It is recommended that implementation of the supporting measures explicitly be tied to the timeframe for completing capital account liberalization and the liberalization plan is kept under review to account for progress in establishing the prerequisites.

48. The authorities should consider liberalizing the capital account vis-à-vis all countries, not just regional partners. Since Uganda’s capital account has already been liberalized, Rwanda plans to lift the controls soon, and Kenya has only a very limited number of controls in place, the opening of Tanzania’s capital account within the EAC effectively implies capital account liberalization vis-à-vis all other countries. The absence of controls in other EAC countries does not allow the enforcement of the remaining controls in Tanzania towards third countries.

C. Pensions and Insurance

49. Tanzania’s pension funds are significant purchasers of long-term financial instruments. The seven funds have approximately 800,000 members and funds under management in excess of T Sh 1,000 billion. The funds cover largely employees within the wider public sector. While membership is mandatory, coverage is low at 40 percent of the formal sector. The investment portfolios of these funds are undiversified and illiquid with a heavy weighting in long-term government securities and commercial real estate. Recent legal changes permit voluntary supplementary schemes, but none yet exist.

50. With the Social Security Regulatory Agency (SSRA) not yet operational, oversight is lacking; the supervisory function should be established without delay. The SSRA should be made operational as soon as possible, including by appointing a Board and a director general. The BOT and the SSRA should enter into a comprehensive MOU that defines the roles, responsibilities, functions, and activities of each agency; in the medium term, the authorities should review and amend the SSRA Act to address a number of significant deficiencies, including by clarifying the respective supervisory functions of the two agencies. In addition, it is recommended that separate laws for each pension fund be replaced by a single law and that existing inconsistencies with the SSRA Act be redressed.

51. Insurance companies are small and write mainly short-term business; the life-segment is underdeveloped. The non-life market is dominated by four companies with an aggregate market share of 52 percent, with the remainder divided among twelve other institutions. The segment is small, with premiums of T Sh 191 billion and a penetration of 0.8 percent of GDP, derived mainly from motor, property and marine businesses. The life segment is underdeveloped with written premium of T Sh 18 billion and insurance penetration of 0.1 percent. Four companies operate in this segment of the market. The government-owned National Insurance Corporation (NIC) has suffered a sharp decline in market share in both markets due to loss in confidence reflecting its inability to pay claims.

52. A number of reforms to the insurance market are recommended:

  • A further increase in the minimum capital requirement for all types of insurers to US$5 million, beyond the currently planned gradual increase to US$1 million.

  • Adoption of minimum disclosure requirements (including for fees and commissions) for insurance brokers, which generate an estimated 60 percent of gross premiums.

  • Short-term resolution of the NIC. The long-standing inability to pay claims of NIC (already noted in the 2003 FSAP) continues to impact confidence and the development of the insurance market.

  • Finalization of arrangements to allow bancassurance, which will improve the distribution channels in the market.

Appendix I. Detailed Summary of Recommendations 1/

article image
article image
article image
article image
Notes:

Based on Aide-Mémoire of FSAP mission, January 2010.

S = short term (3–6 months); M = medium term (1–2 years); L = long term (3–5 years).

Appendix II. Risk Assessment Matrix

article image
article image
article image
article image
article image

Appendix III. Summary of Stress Testing Methodology13

150. This appendix describes the methodology for the stress tests on the Tanzanian banking sector carried out by the FSAP Update mission. The mission applied single factor sensitivity analysis for credit, exchange-rate, interest-rate and liquidity risks to simulate the potential impact of exceptional but plausible shocks. It also simulated the effect of adverse changes in macroeconomic conditions through a multifactor shock.

151. The stress testing exercise suffered from major deficiencies in the data, preventing inclusion of all banks in some stress tests. The macroeconomic scenarios and the sensitivity analysis of interest rate risk could only be done for the 10 largest banks, covering 81 percent of total assets of the Tanzanian banking sector. Stress tests for credit, exchange rate and liquidity risks were carried out for 35 banks. For analytical purposes, the banks were grouped into foreign-owned and domestically-owned institutions, or, alternatively, large, medium and small, and community banks.

152. The stress tests included the major risks faced by banks, as follows.

  • The credit risk tests suffered most from data difficulties. Weak bank-level data on NPLs, lack of longer reliable and consistent time series on NPLs, and the absence of a macroeconomic model to relate loan quality to changes in macroeconomic conditions made a more sophisticated analysis of credit risk impossible. The mission chose therefore to assess general credit risk through a sensitivity analysis that assumed a large but plausible increase in NPLs. To get a sense of the magnitude of a large shock, it looked at individual bank experiences. One bank saw an increase in NPLs in the 12-month period up to end-June 2009 of 90 percent. To assume a general increase of that magnitude appeared unreasonable; to keep the change plausible, one-half of the increase was assumed, amounting to a 45 percent increase in NPLs for all banks. Because a breakdown of NPLs in categories of nonperformance was not available, the exact amount of provisioning required could not be established. To err on the side of caution, the mission assumed a relatively high average rate of provisioning of 50 percent of NPLs. Separately, stress tests of concentration risk assumed that the three largest borrowers of each bank would subsequently default.

  • Exchange-rate risk was tested through applying depreciation and appreciation of the Tanzanian shilling against U.S. dollar by 30 percent. The size of the shock was assumed to be equal to two standard deviations of the times series calibrated using two standard deviations of the time series of the dollar rate during 2008/09.14

  • The test for interest rate risk also used historical data for calibration of the size of the shock. It assumed a return of long-term rates to their maximum value in the period 2005/09, i.e., an increase of 574 basis points. Two assumptions were made about short-term rates: (i) the same increase as long-term rates, i.e., 574 basis points, resulting in a parallel shift upward of the yield curve; and (ii) an increase in short rates to their maximum level in the period 2005/09, i.e., an increase of 1,839 basis points. In the latter scenario, long rates would again increase 574 basis points, with the end result that the yield curve would flatten out considerably. The stress test does not measure capital losses (as most banks hold securities to maturity). Rather, it measures the impact of repricing gaps on income, with the impact measured as a decline in net interest income over one year. With limited maturity data available, the test grouped assets and liabilities in two buckets by maturity: <1 year and >1 year, with the short rate increase applied to the former and the long rate increase to the latter. Then income and expenses over one year were calculated on assets and liabilities, respectively. The difference is the loss, which was subtracted from capital.

  • Liquidity risk was simulated by assuming deposit withdrawals equal to two standard deviations of the quarterly time series of banks’ deposit base over last five years. This assumption yielded a daily withdrawal rate of 34 percent for current deposits and 16 percent for term deposits. The test also assumed that banks would have ready access to 90 percent of liquid assets. Banks needing liquidity beyond this threshold would be considered illiquid and would fail. A second test concerned non-resident deposit withdrawals only. Due to difficulties in obtaining data on the historical volatility of non-resident deposits, and taking into consideration the fact that nonresident deposits are comparatively small and account for only 1.2 percent of total deposits, it was assumed that the shock would see a sudden withdrawal of all non-residetn deposits, while leaving resident deposits unchanged.

  • Two multi-factor shocks were simulated:

    • The first multi-factor shock assumed sudden fall in global commodity prices by 30 percent. This would lead the value of exports to decline sharply, exerting pressure on the exchange rate, causing a rapid deterioration by 30 percent over one quarter. A widening current account deficit and deteriorating macroeconomic fundamentals result in a loss of confidence, and parent banks start to impose stricter limits on short-term funding for Tanzanian subsidiaries, while FDI declines. As a consequence, economic growth falls from 5 percent to 2 percent. The assumed magnitude of the shocks was based on analysis of historical data related to commodity price shocks.15 The shock effects were estimated over two years. The paths of major macroeconomic and market variables under the baseline forecast and the stressed scenario are shown in Appendix Table 1. To estimate the effects of the combined shock on the balance sheets of banks, the effects of the single factor shocks are aggregated.

    • Multi-factor shock 2 assumes an increase in political risk that leads to capital outflows. Withdrawal of deposits of non-residents and loans from parent banks to subsidiaries in Tanzania results in a depreciation of the Tanzanian shilling by 30 percent against the U.S. dollar. In addition, the BOT increases interest rates to prevent further depreciation of the national currency. Short and long-term interest rates rise: as in the single factor shock described above, the yield curve shifts upward by 574 bps. As a result, NPLs increase by 45 percent as in the case of the single factor shock. To estimate the effects of the combined shock on the balance sheets of banks, the effects of the single factor shocks are aggregated.

Appendix Table 1..

Assumptions for Multi-Factor Shock 1

article image
Source: IMF calculations.

153. Because of the short available time series, the multifactor stress tests used a panel data methodology to estimate the relationship between probabilities of default (PODs) in the loan portfolios of banks and various macroeconomic variables. Quarterly data for 2006Q2-2009Q2 on loans and NPLs by sector was obtained for the ten largest banks. Given the absence of frequency PODs, proxy PODs were defined as follows: PODsSector = NPLsSector/Total LoansSector. For econometric calculation purposes, PODs were transformed to logit format to allow for non-linearities to yield the dependent variable ln(PODs/(1-PODs). Finally, the model was specified as follows:

PoDi,t=ci+ αPoDi,t−1+Σn=1kβ F,t  MF,t

where PODsi, t denotes the logit transformed PODs for bank i in quarter t, Ci stands for a fixed effect for bank I, α is the elasticity of the autoregressive term that helps to explain the impact of macro variables on changes in PODs over time, and β denotes the elasticity of the impact of macro factor M at the time period t.

154. The analysis used a Generalized Methods of Moments (GMM) estimation method with random effects. Hausman test results confirmed the existence of random effects in all the three equations. Inclusion of the autoregressive term made the panel dynamic; therefore a panel GMM estimation technique was applied.

155. The initial set of macroeconomic variables included global commodity food, coffee, cotton, gold, and fuel prices as well as domestic variables: GDP growth, a short-term interest rate and the Tanzanian shilling/U.S. dollar exchange rate. The data were converted into logs, except for GDP growth. Panel data analysis showed a significant relationship between PODs and global commodity food prices as well as GDP. All the other variables proved to be insignificant, although their direct impact on PODs might be masked by the quality of historical data.

156. The analysis used separate loan data for six sectors: agriculture, manufacturing, transport, and communication, trade, personal, and other. For manufacturing, transport and communications, and other a robust econometrical relationship could not be established, so that the ad hoc assumption was made that PODs would increase by two standard deviations for each bank. For the other sectors, the estimated equations are as follows:

article image

157. Losses were calculated for each bank and each sector separately. Shocked macro-variables (global commodity food prices and GDP) were inserted into the estimated equations to get shocked PODs. The difference between the current level of PODs (Q2:2009) and their forecasted level was calculated for each bank and sector for each quarter from Q3:2009 to Q2:2011. The difference was multiplied by each bank’s exposure to the respective sector. A loss given default (LGD) ratio of 50 percent was applied and removed from the risk-weighted assets (RWA). Banks were assumed to make provisions equal to the current actual ratio of provisions to NPLs. Net interest income was assumed to decline in proportion to the increase in the ratio of NPLs to RWA.

Appendix IV. Basel Core Principles Assessment—Main Findings

A. Introduction

158. This is a summary of an update assessment that was prepared in the context of the FSAP Update mission conducted during September 9 to 23, 2009. It follows a first assessment of Tanzania’s compliance with the BCPs was done as part of the 2003 FSAP. The update reflects the banking supervision practices of the Bank Supervision Department (BSD) of the BOT as of end-August 2009. The assessors were Mr. Keith Bell and Mr. Richard Hands.

B. Information and Methodology Used for Assessment

159. The assessment is based on several sources: (i) interviews with staff of the BSD; (ii) a review of the legal and regulatory framework, (iii) meetings with senior officials of domestically-owned banks as well as foreign bank subsidiaries incorporated and operating in Tanzania; and (iv) an interview with a partner of a major international accounting firm active in Tanzania. A self-assessment completed by the DBS and dated September 1, 2009 had not been prepared using the structure and methodology recommended in the October 2006 BCPs Methodology document and was of marginal utility to the assessors. The present assessment was performed in accordance with the guidelines set out in the October 2006 document.

C. Institutional and Macro-Prudential Setting—Market Structure

160. The financial services industry in Tanzania comprises the banking sector (i.e., universal and commercial banks), collective investment schemes, insurance business and pension funds. The industry is small but growing, and is dominated by the banking sector, which accounts for some four-fifths of the financial services industry’s total assets. There are 36 licensed banks, a large majority of which are private, and foreign participation is slightly above 50 percent.

161. The banking sector comprises three main segments; large domestic banks, subsidiaries of major international banks, and the smaller—mostly domestic—banks. The three large “domestic” banks (National Bank of Commerce (NBC), National Microfinance Bank (NMB), and Cooperative and Rural Development Bank (CRDB)) account for approximately one-half of the sector’s assets at end-June 2009. Four subsidiaries of major international banks (Citibank, Standard Chartered, Stanbic, and Barclays) take up a further 25 percent, with smaller banks accounting for the remainder. Asset concentration—although declining—is still high, with the seven noted banks accounting for almost three quarters of the total assets of the banking sector.

162. The BOT is Tanzania’s bank supervisory, regulatory, and licensing authority by virtue of the Bank of Tanzania Act, 2006 and the Banking and Financial Institutions Act, 2006 (BFIA). It operates a fairly comprehensive and transparent system in exercising its exclusive right to grant banking licenses and to maintain observance of bank legislation and regulations.

D. General Preconditions for Effective Banking Supervision

163. Effective banking supervision requires a set of preconditions to be in place. These include: (i) sound and sustainable macroeconomic policies; (ii) a well-developed financial infrastructure (legal framework, accounting, and auditing); (iii) effective market discipline; (iv) procedures for efficient resolution of problems in banks; and (v) mechanisms for providing an appropriate level of systemic protection (safety net).

164. The global financial crisis has adversely impacted the otherwise strong macroeconomic performance and the current outlook. Economic growth was expected to slow to 4 to 5 percent in 2009 (from 7.5 percent in 2008) due to a decline in traditional exports and weakening tourism and FDI. Inflation accelerated in 2009, mainly due to increased food and fuel prices. The balance of payments came under pressure as a result of portfolio outflows and a widening current account deficit reflecting continued high import growth. Against this background, the main macroeconomic risks for the banking sector are further deterioration of growth prospects, rising unemployment, prolonged fall in commodity prices, and lower external demand for traditional exports.

165. The legal framework has changed considerably since the 2003 assessment. The Bank of Tanzania Act, 2006 (BOTA’06) and the Banking and Financial Institutions Act, 2006 (BFIA’06) detail the BOT’s functions and powers regarding licensing and supervision of banks and nonbank financial institutions. Specific prudential requirements are contained in the BFIA’06 and its attendant regulations and also in directives and circulars issued by the BOT. The Companies Act 2002 that came into force in March 2006 strengthened aspects of the corporate governance framework, particularly in respect of directors’ duties.

166. Responsibility for promulgation of accounting and auditing standards rests with the National Board of Accountants and Auditors (NBAA). The NBAA has adopted the International Financial Reporting Standards (IFRS) for application for all commercial enterprises effective July 2004. Supervised institutions must appoint annually an external auditor from a register of auditing firms approved for the purpose by the BOT and the audit must be completed in accordance with International Auditing Standards with the auditor to opine as to whether the supervised institution’s financial statements have been prepared in accordance with the IFRS. The NBAA appears not well resourced to monitor and enforce compliance with accounting standards.

167. Under the BFIA’06, the BOT has extensive corrective powers. These include explicit authority to require corrective action plans from banks, including requirements for a capital plan, plans for strengthening a banks’ financial position, and other measures the BOT deems necessary. Yet it is important that these powers be used effectively and in a timely manner, with incipient problems in banks addressed before they cause a breach of prudential limits. The assessors noted that a number of banks appeared in breach of prudential regulations as of the time of the assessment, including three banks with CARs below the required minimum.

168. Under the BFIA’06, there exists a DIF managed by the Deposit Insurance Board (DIB). The Governor of the BOT chairs the DIB and the BOT “make(s) available to the DIB such facilities and the services of such officers as are necessary for the proper and efficient exercise of the functions of the DIB”. Every bank and licensed FI must contribute to the DIF in such annual amount as the DIB determines. In line with the 2003 FSAP recommendations, the DIF has been confined to a pay-box, but the DIB retains extended functions, such as liquidation responsibilities. Liquidation regulations have yet to be completed. Such regulations would allow the DIB to liquidate a failed institution outside of the provisions of the Companies Act, so as to facilitate a prompt and cost effective winding-up of a failed institution.

E. Principle-by-Principle Assessment

169. The DBS has made substantial progress toward establishing its bank supervision at the level of international best practice. Overall, the assessment shows that the DBS has achieved full compliance with 4 of the 30 core principles and sub-principles (Appendix Table 3). It is largely compliant (i.e., has only minor exceptions) with 22 of the Principles and sub-principles and materially non-compliant with the remaining four. In summary, the DBS has in place the fundamentals of a sound supervisory system. The assessments are summarized below.

Appendix Table 2.

Recommended Action Plan

article image
article image
article image
Appendix Table 3..

Summary Compliance of the Basel Core Principles

article image
Notes:

C: Compliant.

LC: Largely compliant.

MNC: Materially non-compliant.

NC: Non-compliant.

NA: Not applicable.