In its 2003 Report on the Observance of Standards and Codes (ROSC) on fiscal policy, the International Monetary Fund (IMF) found that Bangladesh had achieved some advances in its basic fiscal policy transparency practices. These improvements include bringing accounting and fiscal reporting standards closer to the basic requirements of the IMF's fiscal transparency code. Further improvements were noted in the 2005 ROSC update, which noted that many 2003 recommendations by the IMF had been implemented, particularly with regard to fiscal reporting. Reforms elsewhere had been slower, most recently due to the failure to hold elections scheduled for January 2007. However, the 2007 IMF Article IV Consultation reported that the transitional government, put in place following the missed elections, enjoys wide approval for its anticorruption campaign, which has resulted in scores of arrests in both the business sector and within government, and has enabled the resumption of fiscal reform efforts. Nonetheless, the Open Budget Index for Bangladesh compiled by Artiur Rathman and Unnayan Shamannay ranks the Bangladesh budget process at 40%, earning a descriptive evaluation of "minimal" openness.

General Overview

The International Monetary Fund's (IMF) last full Report on the Observance of Standards and Codes (ROSC) covering Bangladesh's fiscal policy transparency was published in 2003, with a 2005 update. The 2003 ROSC asserted that Bangladesh had progressed in achieving transparency, in part through improvements in accounting and fiscal reporting standards. According to the 2003 ROSC, whereas Bangladesh previously had "a highly inadequate and uncoordinated mechanism" for reporting and reconciliation, it had now achieved "a broadly integrated system capable of producing reports that will meet international standards of timeliness, reliability, comprehensiveness, and analytical capacity" (p. 22). According to the 2003 ROSC, this would lead to greater transparency more broadly in fiscal management. At the time of the 2003 ROSC, however, a number of deficiencies were strongly noted. For instance, "Local government is subject to considerable central direction, administrative capacity is weak, corruption is prevalent, and local government data are not systematically reported" (p. 10). Additionally, the ROSC found that even though the government expressed a commitment to decentralization, the direction such reforms should take was still subject to debate, the local governments were handicapped by capacity constraints and a heavy dependence on central government revenues.

A 2005 ROSC Update by the IMF looked at progress achieved by Bangladeshi officials in implementing further improvements, as had been recommended in the original 2003 report. According to the 2005 update, Bangladesh had not only taken steps to improve fiscal reporting reliability, but also had progressed toward effectuating the capacity for medium-term budgeting. However, the 2005 ROSC did caution that "progress on longer-term issues with regard to improving reporting and control over state-owned enterprise reporting and quasi-fiscal activities, and strengthening government watchdog institutions will require continuing sustained effort" (p. 23). The ROSC update was less glowing about other transparency issues that had been first raised in the 2003 report. These included budget presentation and local government reporting requirements. However, the update noted that the Bangladeshi authorities had established a timetable for improvements.

Writing for the International Budget Project's Open Budget Initiative, Dr. Artiur Rathman and Unnayan Shamannay developed an Open Budget Index (OBI) score for Bangladesh of 40%, which translates to a rating of "minimal" openness in the budget process. The OBI score tracks seven key budget documents, the Pre-Budget Statement, the Executive's Budget Proposal, the Citizens' Budget, In-year Reports, the Mid-year Review, the Year-end Report, and the Auditors' Report. Of these, Bangladesh produces five -- it does not produce the Citizens' Budget or the Year-end Report -- but it only makes two documents publicly available: the Executive's Budget Proposal and In-year Reports. Even these two documents, are deemed in need of improvement. The OBI scored the Executive document as providing only 56% of the information needed to permit the public to reach a comprehensive understanding of the government's activities, and the In-year reports were judged to be less than comprehensive. Overall, the OBI suggested that there was substantial room for improvement in rendering the Bangladeshi budget process more open.

The 2007 IMF Article IV Consultations noted that, with the failure to hold elections scheduled for 2007, a transitional government was installed to serve until new elections could be held. The report stated that the transitional government "enjoyed widespread support for its anticorruption campaign" (p. 4), which has resulted in the arrest of "scores" of individuals both from the business sector and from high levels of the government and has enabled previously stalled reform initiatives to be put in motion once again. For instance, the process of corporatizing the nationalized commercial banks has resumed. However, the report added that the structural reform benchmarks "related to improving coordination between the large taxpayer units and the... [assignment of] unique numbers to large taxpayers were not met" (p. 6). The report noted that pending resolutions to more completely separate the judiciary from the executive branch were adopted, electricity tariffs and fuel prices were raised on schedule, and a committee was formed to conduct regular reviews of fuel prices and recommend adjustments.

The 2007 report lauded the transitional government's success in maintaining broad fiscal discipline, while cautioning that revenue shortfalls contributed to a generally weak fiscal performance. The authorities agreed with the IMF's position, announcing their intention to improve public financial management, explore the possibilities of revenue reforms, continue the restructuring and/or privatization of state-owned enterprises, and take other steps designed to enhance fiscal policy transparency. To these ends, the IMF recommended a complete overhaul of Bangladeshi tax law, and offered to provide technical assistance.

Bangladesh is not a subscriber to the IMF's Special Data Dissemination Standard (SDDS), but has participated in the less rigorous General Data Dissemination System (GDDS) since 2001, as is disclosed on the GDDS website. Both the website and the Statistical Appendix (Appendix IV) of the 2007 IMF report note that "macroeconomic data are adequate for surveillance, despite major shortcomings in national accounts, government finance statistics, and balance of payments" (p. 11).

The Principles

Clarity of roles and responsibilities.

EN

The IMF's 2003 ROSC reported that the Bangladesh Constitution of 1972 clearly defines the various fiscal roles and responsibilities held by the executive, legislative and judicial branches, but added that "many aspects of the relationship need strengthening and are still evolving in practice" (p. 11). Articles 81 through 92 of the Constitution set forth the basic principles governing fiscal management. The office and role of the Comptroller and Auditor-General (C-AG) are laid out in Articles 127 through 132, wherein the C-AG is granted the function of government auditor. This function is further strengthened by provisions of the Additional Functions Act of 1974, whereby the C-AG was given the accounting function, as well. (This latter function has since been transferred to the MoF's Finance Division, according to the 2003 ROSC.) The ROSC noted that public perception viewed the executive and legislative branches of the government to be essentially adversarial. According to the ROSC "parliamentary consideration of economic reforms and development of effective watchdog institutions remains weak" (p. 12).

Although the ROSC found that the Bangladeshi definition of general government was in broad conformity to international practice, it cautioned that "the boundaries between general government, the rest of the public sector, and the rest of the economy are blurred in practice" (p. 5). There is heavy government involvement throughout the economy, and its participation is nontransparent. Involvement with the enterprise sector is carried out through a complex arrangement that includes not only direct subsidies but also quasi-fiscal activities. There is inadequate reporting of both direct and indirect sources of support in this sector, compromising the assessment of fiscal policy sustainability. The ROSC adds that "the lack of clarity of managerial roles between commercial activity and provision of subsidized services or goods to the public is a major factor contributing to widely acknowledged failures of management, corruption, and poor industrial relations in these sectors" (p. 6). Government interventions motivated by the desire to achieve social and economic objectives are common, but a clear statement of these objectives is generally omitted from the budget. The impact on the budget, according to the 2003 ROSC, "is both substantial and, to a large extent, unpredictable, and the extent to which (implicit) policy objectives have been achieved is not monitored or reviewed" (p. 7).

At the time of the 2003 ROSC, the government still maintained extensive, if not comprehensively reported, equity and financial assets, but had announced plans to reduce its state-owned manufacturing enterprises. Some of these units were slated to be closed down, but others were due for privatization. The privatization process was deemed by the ROSC to be transparent but lacking in coordination. The ROSC found more extensive problems with regard to local government. According to the report, "local government is subject to considerable central direction, administrative capacity is weak, corruption is prevalent, and local government data are not systematically reported" (p. 10). Decentralization of government functions remains a priority for the government, but there are capacity constraints inhibiting the implementation of such reforms, and at the time of the ROSC there was no political consensus on just what reforms were best suited to the issue.

The ROSC also noted that there is no legislative basis to assure that monitoring and reporting are carried out with sufficient regularity. Although tax regulations are published, they are nonetheless subject to the discretion of tax officials due to inadequate oversight. Finally, the ROSC reported that there is in place a code of ethics for public sector employees. This, too, is inadequately enforced. According to the ROSC "the widespread prevalence of nontransparent practices in the public sector" (p. 14) has contributed to the ineffectiveness of the code of ethics and asset-disclosure rules in reducing public corruption.

Open budget processes

ID

Writing for the International Budget Project's Open Budget Initiative, Dr. Artiur Rathman and Unnayan Shamannay developed an Open Budget Index (OBI) score for Bangladesh of 40%, which translates to a rating of "minimal" openness in the budget process. The OBI score tracks seven key budget documents, the Pre-Budget Statement, the Executive's Budget Proposal, Citizens' Budget, In-year Reports, Mid-year Review, Year-end Report, and Auditors' Report. Of these, Bangladesh produces five -- it does not produce the Citizens' Budget or the Year-end Report -- but it only makes two documents publicly available: the Executive's Budget Proposal and In-year Reports. Even these two documents, however, are deemed in need of improvement. The OBI scored the Executive document as providing only 56% of the information needed to permit the public to reach a comprehensive understanding of the government's activities, and the In-year reports were judged to be less than comprehensive. Overall, the OBI suggested substantial room of improvement in rendering the Bangladeshi budget process more open.

Bangladeshi fiscal policy focuses primarily on the central government's overall deficit, according to the IMF's 2003 ROSC. The ROSC found that "the annual budget process is clearly defined," but cautioned that it "lacks an open and clearly disciplined priority-setting mechanism, particularly for developments" (p. 16). Spending and revenue priorities are not subject to an adequate overall review, and there is no analysis of risk sensitivity or of the assumptions that underlay the budget. There are only weak internal control and audit mechanisms. Nearly all ministries lacked an internal audit function. On the other hand, the ROSC did note that the majority of non-financial public enterprises did maintain internal audit departments. Even here, however, there was little by way of systematic review of departmental effectiveness, nor were there adequate training programs in place for the audit staff. At the time of the 2003 ROSC, the time lag for the production of audited final accounts was at least 12 months following the fiscal year in question. On a more positive note, the ROSC found significant improvements in the reliability and timeliness of the accounting system, and added that there were plans to begin generating a mid-year fiscal review. Still, the ROSC warned that in-year reports on the central government's budget outturns were not yet complete or accurate. Also lacking in budget and final accounts documents are statements of the accounting policy used in their preparation.

In 2005, the IMF updated the fiscal transparency ROSC, and noted that the Financial Management Reform Program has yielded improvements in fiscal reporting and budget preparation. Still in need of attention, however, is the way in which project funding from foreign donors is monitored and reported. The authorities must work with donors to consolidate development project reporting. According to the ROSC update, "development of capacity in this regard is central to a sustained reform of the accounting and fiscal reporting system. (p. 3-4)

The 2005 ROSC update announced that Bangladesh had, for the first time, approached the principles of the Government Finance Statistics in its budget preparation process for that year. The presentation of the budget to parliament that year also included a statement of the macroeconomic assumptions that guided the budget process, and the budget document contained a summary statement of those assumptions, as well as extensive clarifying notes and explanations of how the budget figures were derived. Even greater detail and more explicit budget analysis was planned for the 2006 budget, and the 2007 budget was expected to contain a statement of tax spending and contingent liabilities, as well. However, the 2005 ROSC update also disclosed that the reporting of quasi-fiscal activities had seen only little improvement, and substantial reform was yet required in reporting on the public enterprise sector. Supervision of the nationalized commercial banks by the central bank is undergoing strengthening, and the government is moving forward with its plans to close down state-owned enterprises and to reform the nationalized commercial banks. Future budgets are expected to include an annex that explicitly deals with the costs and risks of quasi-fiscal activities.

Public availability of information.

ID

Bangladesh is not a subscriber to the IMF's Special Data Dissemination Standard (SDDS), but has participated in the less rigorous General Data Dissemination System (GDDS) since 2001, as is disclosed on the GDDS website. Both the website and the Statistical Appendix (Appendix IV) of the 2007 IMF report note that "macroeconomic data are adequate for surveillance, despite major shortcomings in national accounts, government finance statistics, and balance of payments" (p. 11).

The IMF's 2003 ROSC reported that the Bangladesh budget process comprehensively covers the fiscal activity of the central government, providing both the current year's major fiscal aggregates as well as revised estimates for the previous year. The compilation of central government operations data is prescribed by the Constitution, the Appropriations Act of 1962, and the Additional Functions Act of 1974. The IMF's GDDS website notes that the text of these laws can be found in the "Bangladesh Gazette." The GDDS website also notes that notice of the release of the budget statement is provided about one week in advance. The ROSC cautioned, however, that local government operations data are not included in the budget. Also excluded from the budget are estimates of tax expenditures, quasi-fiscal activities, and contingent liability data. Neither domestic debt nor financial asset data are reported. As of the 2003 ROSC, Bangladesh did not yet achieve timeliness in producing comprehensive data on the central government's budget outturns. This is attributed to "extensive delays in identifying government transactions and problems with reconciliation of fiscal data with information from bank accounts" (p. 15). The ROSC added that the Bangladeshi fiscal authorities had yet to formally commit to regularly publishing fiscal data. At the time of the ROSC, although Bangladesh was already a participant in the IMF's GDDS, Bangladeshi fiscal data did not meet the recommended standards of timeliness, comprehensiveness, and reliability. A series of ongoing reform projects however, include one in which the Finance Division of the MoF began to issue quarterly reports disclosing the fiscal position of the central government. These, in turn, are derived from the monthly Comptroller General's Administration (CGA) reports and the MoF's Economic Relations Department's (ERD) external financing reports. The CGA monthly reports and the Finance Division's quarterly publications are available to the public on the MoF website.

The 2003 ROSC found that Bangladesh has progressed on basic transparency issues. It has improved its accounting and financial reporting standards, moving closer to compliance with the fiscal transparency code, at least with regard to the basics. Similarly, the CGA has made improvements in reconciliation reporting, such that the processes "have been transformed from a highly inadequate and uncoordinated mechanism to a broadly integrated system capable of producing reports that will meet international standards of timeliness, reliability, comprehensiveness, and analytical capacity" (p. 22). The ERD compiles external debt data and produces an annual publication entitled "Flow of External Resources into Bangladesh." This document accompanies the budget when it is presented to parliament, and is released to other interested parties on request. It is not, however, subject to simultaneous public release, according to the IMF GDDS website.

In 2005, the IMF published a ROSC fiscal transparency update, which reported that continued effort must be expended to improve data availability and to more fully integrate state-owned enterprise and nationalized commercial bank data into the budget process. According to the update, addressing the non-transparency and lack of performance reporting in these two sectors "should be given the highest priority for the immediate future" (p. 4).

Independent assurances of integrity.

ID

The IMF's 2003 ROSC noted that there are no publicly available indicators according to which budget data reliability can be assessed. According to the report, budget data reliability would be strengthened if the assumptions that underlay estimates of revenues and expenditures were published. Similarly, statements of the policy guiding accounting practices should be made an integral part of the budget and final accounts, which was not yet the case. The ROSC added that further improvements were needed in the area of accounts reconciliation and fiscal reporting, although some progress had been achieved in those areas covered by the CGA. Also, an interdepartmental Task Force on Monitoring Macroeconomic Data was in place to find ways to introduce improvements in the reconciliation of data regarding foreign-financed projects.

The ROSC found that the external audit function is located within the executive branch, in the Ministry of Finance, but enjoys functional independence. The legal mandate for the Auditor General is found in the Bangladeshi Constitution, Article 128 (4) of which specifies that the AG is free from external control during the conduct of his responsibilities. The Constitution provides for the Auditor General to be a presidential appointee, although the IMF noted that the Finance Division participates in the selection and removal processes. Further, the IMF reported that the appearance of executive branch supervision appears to limit the external audit's functional independence. The ROSC called for the provision of greater audit capacity, noting that only 2,500 of the C-AG's 3,500 positions were filled, and that the higher level offices suffered a disproportionate lack of staffing. Even though there are manuals and updated audit codes, the lack of sufficient staff, along with insufficient training, undermines the ability of the C-AG to carry out its mission. In practice, the 2003 ROSC found that audits gave "too little emphasis to materiality of findings and provides instead detailed reports on financial and regulatory compliance without assessing systemically important deficiencies" (p. 22). Plans were in place to improve training as well as to increase capacity, independence, and effectiveness.

Also deficient, according to the 2003 ROSC, were follow-up procedures for external audits. Part of the problem is attributed to parliamentary delays in dealing with audit recommendations. Contributing to the difficulty is the fact that audit reports tend to lack focus. The ROSC also noted that there was no external assessment of the macroeconomic assumptions employed in fiscal policy. When such assumptions are presented, they are neither comprehensive nor systematic enough to sustain outside scrutiny. Finally, the ROSC reported that there was an increased emphasis on improving reconciliation practices by governmental departments and revenue agencies.

The 2005 ROSC Update noted a number of steps taken since the original ROSC that aimed at improving Bangladesh's performance in regard to this principle. For one, an Economic Monitoring Committee (EMC) was created and charged with the oversight of fiscal reporting. The Update found that the EMC's reports were consistently timely, being produced on a monthly basis, six weeks following the month under review. The Financial Management Reform Project has made the inclusion of reconciliation statements a routine part of its own reports, which has led to a reduction in the use of suspense balances. The ROSC Update adds that government accounts data beginning in fiscal year 2002 employed the format of the Government Finance Statistics, 2001 edition. These data appear on the IMF's GDDS website. The 2005 ROSC further notes that the Ministry of Local Government and Rural Development is also working toward greater accountability and transparency. To this end, the ministry is instructing "all city corporations and municipalities to prepare annual reports and publish audited financial statements" (p. 3). Local government finance and audit staffs are also undergoing training and city corporations are beginning to adopt accrual accounting methods.