Bangladesh: Detailed Assessments Using the Data Quality Assessment Framework (DQAF)

This Report on the Observance of Standards and Codes (ROSC) data module provides an assessment of Bangladesh’s macroeconomic statistics against the recommendations of the General Data Dissemination System (GDDS) complemented by an assessment of data quality based on the IMF’s Data Quality Assessment Framework. With some exceptions, the coverage, periodicity, and timeliness of macroeconomic statistics meet the GDDS recommendations. Guided by GDDS plans for improvement, the Bangladesh Bureau of Statistics, the Ministry of Finance, and the Bangladesh Bank have demonstrated commitment to continuously improve the compilation and dissemination of macroeconomic statistics.

Abstract

This Report on the Observance of Standards and Codes (ROSC) data module provides an assessment of Bangladesh’s macroeconomic statistics against the recommendations of the General Data Dissemination System (GDDS) complemented by an assessment of data quality based on the IMF’s Data Quality Assessment Framework. With some exceptions, the coverage, periodicity, and timeliness of macroeconomic statistics meet the GDDS recommendations. Guided by GDDS plans for improvement, the Bangladesh Bureau of Statistics, the Ministry of Finance, and the Bangladesh Bank have demonstrated commitment to continuously improve the compilation and dissemination of macroeconomic statistics.

Acronyms

1993 SNA

1993 System of National Accounts

ABD

Accounts & Budgeting Department, BB

BB

Bangladesh Bank

BBS

Bangladesh Bureau of Statistics

BPM5

Balance of Payments Statistics Manual, fifth edition

CGA

Comptroller for Government Accounts, MOF

DQAF July 2003

Data Quality Assessment Framework, July 2003 version

DSBB

Dissemination Standards Bulletin Board

GDDS

General Data Dissemination System

GDP

Gross Domestic Product

GFS

Government Financial Statistics

GFSM 1986

A Manual on Government Finance Statistics, 1986

GFSM 2001

Government Finance Statistics Manual 2001

IMF

International Monetary Fund

ITRS

International Transactions Reporting System

MFSM

Monetary and Financial Statistics Manual

MOF

Ministry of Finance

MOP

Ministry of Planning

ROSC

Report on the Observance of Standards and Codes

SD

Statistics Department, BB

SDDS

Special Data Dissemination Standard

TFQA

Task Force on the Quality of Accounts

Detailed Assessment Using the Data Quality Assessment Framework (DQAF)

The following detailed information on indicators of statistical practices in the areas of the national accounts, prices, government finance, money and banking, and balance of payments statistics was gathered from publicly available documents and information provided by the Bangladesh officials. This information, which is organized along the lines of the generic DQAF (see Appendix II), was used to prepare the summary assessment of data quality elements, based on a four-part scale of observance, shown in Bangladesh’s Report on the Observance of Standards and Codes (ROSC)—Data Module.

I. National Accounts

0. Prerequisites of quality

0.1 Legal and institutional environment

0.1.1 The responsibility for collecting, processing, and disseminating the statistics is clearly specified

The Bangladesh Bureau of Statistics (BBS) is the national statistical office of the Government of Bangladesh and operates under the guidance of the Ministry of Planning (MOP). The responsibility of the BBS to compile and disseminate national accounts statistics derives from the government “Allocation of business among different ministries and division” (Schedule 1 of the Rules of Business) from 1996. These rules of business cover only limited aspects of the statistical activities, mainly the data collection activities relating to special surveys and censuses (such as the population census, the agricultural census, the manufacturing census, and the household income and expenditure survey). No statistical legislation clearly defines the activities of the BBS as the national agency responsible for compiling and disseminating national accounts statistics. However, the official statute of the BBS as the agency compiling national accounts statistics is recognized.

0.1.2 Data sharing and coordination among data-producing agencies are adequate

No documented and established procedures ensure timely flow of data for the national accounts. Data sharing and coordination are based on goodwill among the data producers. Even within the BBS, there are no set arrangements for data exchange among the statistical wings. Data from other statistical wings are made available to the National Accounting Wing with long delays.

A serious weakness in the compilation of the national accounts is that all information exchange between agencies is mainly paper-based. Besides, contacts with the data providing agencies follow a rigid process, which is not conducive to a quick resolution of data problems. The National Accounting Wing has to send written requests for data to other agencies and to follow up by phone or by fax. In addition, the national accounts staff have to wait for the availability of final publications.

The absence of legal mandate that establishes the authority of an agency to coordinate statistical services has contributed to weak coordination between the data producing agencies in Bangladesh. There are no regular meetings between the data producing agencies, and issues such as response burden or duplication of efforts have not been addressed. The role of the National Statistics Council, the highest policymaking body in the field of statistics on a national level, appears to be limited to approving national censuses and surveys, or discussing methodological issues. The recently created Task Force on Monitoring of Economic Data, chaired by the Joint Secretary of the Budget, has also a limited role in discussing the consistency of the national accounts with other macroeconomic data. The authorities have plans to create an Advisory Committee on national accounts, aimed at coordinating the efforts of agencies in improving the quality of the national accounts. The Committee will be chaired by the Minister of Finance, with representatives from data producing agencies, academia, researchers, and other users of national accounts data.

0.1.3 Individual reporters’ data are to be kept confidential and used for statistical purposes only

The BBS treats individual data as strictly confidential even in the absence of a legal provision mandating confidentiality. Statistical Acts cover specific data collection, relating to the census of manufacturing industries, the population census, and the agricultural census.

The BBS informs the respondents of its obligations to preserve confidentiality when distributing the statistical questionnaires. The questionnaires assure the respondents that the information they provide will be used only for statistical purposes and will not be published in any form that discloses the identity of the unit.

No written regulations prevent disclosure of individual reporters’ data. However, access to individual data is restricted only to compilers and supervisors. All computers are password-protected. The BBS staff reviews the data before dissemination to make sure there is no indirect disclosure of individual data.

0.1.4 Statistical reporting is ensured through legal mandate and/or measures to encourage response

Statistical reporting is not ensured through legal mandate. Data collection is based on voluntary responses. The Statistical Acts for censuses provide minimal penalty for noncompliance (the equivalent of US$3). In practice, penalties for noncompliance are never imposed.

0.2 Resources

0.2.1 Staff, facilities, computing resources, and financing are commensurate with statistical programs

Alhough staffing for national accounts compilation is adequate, computing resources and budget for statistical surveys are not adequate.

The staffing of the National Accounting Wing is adequate with a staff of 11 statistical officers and 20 supporting officers. The current data compilation practice of the national accounts is extremely resource consuming, most of the source data come in hardcopy format and the entry is done manually at the National Accounting Wing. For example, the consolidation of the government finance statistics is done by the national accounts staff using hardcopy budget statements received from numerous units of government. The training received is generally adequate. Most of the staff have university degrees. Training in national accounts has been provided through courses in Bangladesh organized by donors, mainly the Asian Development Bank. Two staff members have been trained in IMF training courses in national accounts and one staff has attended a training course at the U.S. Bureau of Economic Analysis.

Staff who join the Bangladesh Civil Service tend to remain in the service until retirement, and this is true for the BBS. Staff turnover is low. Job rotation is implemented internally by the BBS management after five or six years of service of a statistical officer. Salary levels for civil servants in Bangladesh are perceived by the BBS staff to be competitive.

The computing resources for compiling national accounts are not adequate. Only the statistical officers have personal computers, which the technical staff are allowed to use on an ad hoc basis. Computers were received from donors for carrying out donor-financed surveys. Most of the computers are with low capacity processors. The National Accounting Wing does not have a central processing unit and is not connected to other data processing units of the BBS. Staff have no access to Internet. Management has not reviewed the information technology needs of the national accounts compilers. There is no special software to be employed in the national accounts compilation.

The BBS has a limited budget for financing statistical surveys needed for the national accounts compilation. The only surveys financed from the BBS budget are the census of manufacturing industries and the agricultural census. All other surveys undertaken currently by the BBS (such as the household income and expenditure survey, and the labor force survey) are financed from donor funds.

0.2.2 Measures to ensure efficient use of resources are implemented

No internal process measures resources used to compile the statistical series or compares the resource usage of the statistical program vis-à-vis other statistical programs. There is no unit in the BBS to review the efficiency of the work process. In practice, the BBS has no annual plan for statistical surveys, and the main source data for the national accounts are the census surveys. The heavy reliance on census data is costly and time-consuming. Moreover, there is no guarantee that the data would be of superior quality to what may be obtained from sample surveys.

The National Accounting Wing seeks outside expert assistance to evaluate methodologies and compilation methods for the national accounts.

0.3 Relevance

0.3.1 The relevance and practical utility of existing statistics in meeting users’ needs are monitored

Periodic users’ surveys are not conducted. However, in December 2004, the national accounts staff organized their first meeting with users from the public and the private sectors to assess the national accounts compilation and dissemination process and seek suggestions for improvement. Participation of national accounts compilers in international statistical meetings and seminars organized by regional and international organizations is limited.

0.4 Other quality management

0.4.1 Processes are in place to focus on quality

The management recognizes quality as an important statistical principle. Recent initiatives are the participation in the IMF’s General Data Dissemination System (GDDS) and in the data module ROSC. Improvement of quality is the objective of the multiyear project for strengthening the national accounts, which started in 1996 with the support from the Asian Development Bank. However, a detailed plan for implementing the 1993 System of National Accounts (1993 SNA) and improving quality of the current national accounts estimates does not exist.

0.4.2 Processes are in place to monitor the quality of the statistical program

Although the BBS has not established an explicit program of quality awareness, many elements of such a program are already in place, especially in monitoring data collection and processing. The quality of data is monitored from the time data are collected in the field (through inspections, cross-checking, etc.) until the compilation and dissemination stage.

0.4.3 Processes are in place to deal with quality considerations in planning the statistical program

The management recognizes the alternatives among the dimensions of data quality. An example is the decision to release projected estimates of the national accounts, recognizing the risk of being criticized when the final data are released. However, there are no trade-offs between dimensions of quality, such as timeliness and accuracy, because the timeliness of the national accounts estimates is not predetermined by an advance release calendar.

1. Assurances of integrity

1.1 Professionalism

1.1.1 Statistics are produced on an impartial basis

There is no law or other legal provision that addresses the need for professional independence of the BBS. As previously mentioned, the BBS is fully integrated within the MOP. The National Accounting Wing provides assurances that it resists political interference in the national accounts estimates. The national accounts aggregates are compiled on an impartial basis.

The procedures for the collection and compilation of national accounts statistics are governed by technical considerations. Professionalism is promoted by the publication of methodological papers and special studies. However, this objective could be diluted by a promotion policy that is based on seniority measured by the length of the service.

1.1.2 Choices of sources and statistical techniques as well as decisions about dissemination are informed solely by statistical considerations

The choice of statistical sources is restricted by the limitations of the data collected by other statistical wings of the BBS and public agencies. There is no interference of any type in the choice of statistical techniques for compiling the national accounts.

Publication of the data by the BBS is always accompanied by some methodological explanations. An extensive written description of methods is being developed within the framework to strengthen the national accounts project.

1.1.3 The appropriate statistical entity is entitled to comment on erroneous interpretation and misuse of statistics

The National Accounting Wing is entitled to comment on and respond to any misused or misinterpreted data. The national accounts staff scan the major media releases on a daily basis. Only a few cases of misinterpretation have been recorded so far, and the staff have provided explanations.

1.2 Transparency

1.2.1 The terms and conditions under which statistics are collected, processed, and disseminated are available to the public

The terms and conditions under which the BBS works are laid down in the Rules of Business and the Statistics Acts for data collection. The documents are freely available to the public. Further explanations on the terms and conditions are found on the IMF’s Dissemination Standards Bulletin Board (DSBB).

1.2.2 Internal governmental access to statistics prior to their release is publicly identified

The dissemination of the national accounts estimates is always cleared by the Minister of Planning. This procedure is disclosed in the IMF’s DSBB, but not in the BBs’s website or printed publications. Prior access to data combined with the absence of a preannounced dates of release has already affected the public’s perception of the agency’s transparency.

1.2.3 Products of statistical agencies/units are clearly identified as such

Products of the BBS are clearly identified as such by its name and logo. The national accounts tables disseminated in other agencies’ publications identify the BBS as the source. However, the BBS does not mention in its publications that attribution should be made to it when its statistics are used or reproduced.

1.2.4 Advanced notice is given of major changes in methodology, source data, and statistical techniques

Major changes in methodology and source data are announced in advance, for example, the implementation of a program for changing the base year. Users of statistics are informed through press releases and publications when new data are disseminated. In particular, when a new base year is published, advance public notice is given to explain the methodological improvements and the differences between the old series and the new series.

1.3 Ethical standards

1.3.1 Guidelines for staff behavior are in place and are well known to the staff

Staff behavior is guided by the administrative guidelines of the civil service, which are made available to staff upon assumption of duty.

2. Methodological soundness

2.1 Concepts and definitions

2.1.1 The overall structure in terms of concepts and definitions follows internationally accepted standards, guidelines, or good practices

The 1993 SNAis largely followed as a general framework for compiling national accounts statistics.

2.2 Scope

2.2.1 The scope is broadly consistent with internationally accepted standards, guidelines, or good practices

The minimum requirements for the scope of national accounts tables and accounts, established by the Intersecretariat Working Group on National Accounts (ISWGNA), are not met. The BBS compiles and disseminates the following national accounts and aggregates:

  • annual value added and gross domestic product (GDP) at current and constant (1995/96) prices by activity; and

  • annual expenditure of GDP at current and constant (1995/96) prices.

The BBS compiled an experimental sequence account by institutional sector for only one year (1995/96); however, the ISWGNA requires annual compilation. Further, the BBS does not compile and disseminate quarterly GDP estimates or annual supply and use tables as recommended by the ISWGNA.

The national accounts cover the economic territory of the country. The residence criterion is defined in accordance with the 1993 SNAand the Balance of Payments Manual, fifth edition. The following are included:

  • territorial enclaves such as Bangladesh embassies in the rest of the world;

  • bonded warehouses and free trade zone activities; and

  • workers who work only part of the year in another country.

However, since the Bangladesh Bank (BB) does not include the offshore banks in its depository corporations’ survey, these banks are respectively not included in the national accounts estimates.

The production boundary is in accordance with the 1993 SNA. In particular, the scope of output measurement includes, among other things, the following items:

  • own-account production of goods for own-final consumption;

  • output of goods for own-account fixed-capital formation;

  • research and development on own account;

  • mineral exploration; and

  • production of computer software, entertainment, literary and artistic originals.

Illegal activities are not covered in the scope of output measurement.

Among tangible assets, the defense related assets that could be used for civilian purposes, the valuables, the agricultural work-in-progress, and historical monuments are not at present in the scope of the assets boundary.

Among intangible assets, patented entities, leases, and other transferable contracts are not included in the assets boundary.

Deviations from the 1993 SNAare kept under review, which may lead to future revisions.

2.3 Classification/sectorization

2.3.1 Classification/sectorization systems used are broadly consistent with internationally accepted standards, guidelines, or good practices

The classification and sectorization used in the compilation of the national accounts broadly conform with internationally recommended standards.

The 1993 SNA is in principle followed to classify institutional units. The BBS data on general government sector identify the various levels of government such as central, city corporations, district (zila parished), municipality, and county (union parished) government. In addition, the national accounts staff make adjustments to the central budget data received from MOF to classify some units (the railways, post services, telegraph services) in the nonfinancial corporations sector.

The BBS has implemented the following classifications for national accounts that consider the specific recording requirements of the Bangladesh statistics:

  • the Bangladesh Standard Industrial Classification to classify activities, consistent with the third revision of the International Standard Industrial Classification; and

  • the Central Product Classification to classify products.

Current expenditures of general government are classified using a national classification, broadly consistent with the Classification of the Functions of Government.

Data for individual consumption expenditure from the Household Income and Expenditure Survey (HIES) are classified according to a national expenditure classification, which is not fully consistent with the Classification of Individual Consumption by Purpose (COICOP).

Foreign trade transactions are classified according to the Harmonized System (HS).

2.4 Basis for recording

2.4.1 Market prices are used to value flows and stocks

The valuation rules used for recording flows are generally in accordance with the 1993 SNA. Market output is valued at producers’ prices. Output for own use is valued at equivalent market prices. The valuation of intermediate consumption is at purchasers’ values, that is, including excise taxes and nondeductible value added tax. The deductible part of the value added tax is excluded from the valuation of final uses. GDP is presented at market prices. Corrections are made when transfer prices are detected. Information on insurance and freight for merchandise imports is not available. Imports of goods are recorded at cost, insurance, and freight (c.i.f), and exports at their free on board (f.o.b.) values, while the 1993 SNArecommends f.o.b. valuation for both categories. The BBS does not convert transactions in foreign currency into Bangladeshi taka. Data on imports are converted into local currency by the BB. The conversion is done using the midpoint exchange rate prevailing in the market at the time the transactions take place.

The national accounts are compiled according to the fiscal year of the government, which is July to June.

2.4.2 Recording is done on an accrual basis

Most transactions are recorded on an accrual basis, but the government transactions and the rest of the world transactions are recorded on a cash basis. Work-in-progress for construction is recorded in the period it is produced, but agriculture work-in-progress is not estimated, as recommended by the 1993 SNA. Instead, the production of crops is recorded in the period of harvest.

2.4.3 Grossing/netting procedures are broadly consistent with internationally accepted standards, guidelines, or good practices

Transactions between establishments within the same enterprise are recorded on a gross basis.

3. Accuracy and reliability

3.1 Source data

3.1.1 Source data are obtained from comprehensive data collection programs that take into account country-specific conditions

The BBS does not have a comprehensive data collection program to support the compilation of the national accounts statistics. The estimates are based on census surveys (most of which were conducted in 1995/96), administrative records, and some ad hoc surveys used for indirect estimation.

No single, reliable, and up-to-date statistics register is used as a basis for the statistical surveys. Instead, the different statistics wings of the BBS maintain their own lists of establishments, but these lists do not contain information on large producing units. The lists are based on outdated census surveys. For instance, the sample frame for the manufacturing industries is based on the economic census of 1986.

The main basic sources for the annual accounts estimates are summarized in the table below:

Table 1.

Main Sources for the Annual National Accounts (final 2002/03 estimates)

article image
article image

As mentioned in 0.2.2, the national accounts rely heavily on census data, which are costly, time-consuming, and outdated. Most census surveys are more than seven years old. The ad hoc surveys are also outdated. The main problem of the source data for the national accounts is that direct derivation of all aggregates (output, intermediate consumption, fixed capital formation, changes in inventories) is only possible from the census surveys. The ad hoc surveys and the administrative source data allow the estimation of sales only, which oftentimes are based on volume indicators.

The BBS conducts a HIES, but not annually. The latest HIES was undertaken in 2000. The survey sample is based on the Population Census of 1991. A two-stage stratified random sampling technique was used. The sample frame for the urban areas was different from that used for the rural areas. In the first stage, the enumeration blocks and villages were considered the primary sampling units (PSU). The PSUs were selected from 14 different strata—five rural and nine urban. In the second stage, ten households were selected from each PSU through systematic random sampling. A total of 7,440 households were selected in the sample. The interviewers were provided with laptop computers for data entry in the field. The response rate from the households was high. The survey covered all urban and rural areas as well as all income levels.

Data on government expenditures are obtained from the MOF and other respective levels of government as follows:

  • Central government—MOF

  • City corporations—City finance departments (6)

  • District—District governments (64)

  • Municipality—Municipality governments (260)

  • County—County departments (4484)

Detailed MOF data are available to measure output, intermediate consumption, fixed-capital formation, and final consumption expenditure of government services. No breakdown exists by type of the defense related expenditures. Extrabudgetary funds are included in the central government budget. Annually, the National Accounting Wing collects data from all units of local government, which is time-consuming and costly.

The balance of payments statistics are received from the BB.

The price statistics compiled by the BBS are the consumer price index (CPI), the producer price index (PPI), the wholesale price index (WPI), and the imports and exports price indices (IEPI).

3.1.2 Source data reasonably approximate the definitions, scope, classifications, valuation, and time of recording required

The concepts and the definitions used in the census surveys and the HIES are consistent with national accounts methodology, because these surveys were developed with input from the national accounts staff. As mentioned in 3.1.1, the data from administrative sources and the ad hoc surveys measure mostly trends in revenues from sales.

There are some concerns about the methodology of the price statistics compiled by the BBS. The base year of the indices is the following: CPI and PPI—1995/96, WPI—1969/70, and IEPI—1988/89. Apart from the outdated base year, the IEPI is a unit value index and most likely measures not only the price changes but also changes in the product mix, because it is compiled for all HS groups (including the nonhomogeneous ones). The PPI is also a unit value index, instead of a transaction price index.

3.1.3 Source data are timely

Most of the source data for the national accounts are not timely. The National Accounting Wing receives data from other statistical wings of the BBS only after the data have been officially published. The data received from the MOF are also not timely. In fact, the MOF compiles two sets of budget data—revised budget data (based on the actual data for the first nine months of the financial year) and actual data. The national accounts compilers use the revised budget data instead of the actual data, because the actual data of the central government are available two years after the end of the financial year. The monetary and the balance of payments data received from the BB are timely. Administrative data from public enterprises are received with delays, owing to the auditing process.

3.2 Assessment of source data

3.2.1 Source data—including censuses, sample surveys and administrative records—are routinely assessed, e.g., for coverage, sample error, response error, and nonsampling error; the results of the assessments are monitored and made available to guide statistical processes

The source data are collected by the 507 county (thana) statistical offices, and staff from these offices perform manual validation. The hardcopy questionnaires are sent to the 23 regional offices, where further manual validation is performed again. Finally, the hardcopy reports are sent to the headquarters. The first software validation of data takes place in the Central Data Processing Unit of the BBS. If data are considered inaccurate, the headquarter staff return the questionnaire to the regional/county office for corrections.

Source data validation at headquarters makes use of the software and hardware available. High-value transactions are confirmed with respondents. Information on sampling errors is available for the HIES; for all other surveys, there is information only about the nonsampling errors. Source data are analyzed for underreporting and temporal consistency. Adjustments to source data are done for nonsampling errors.

The BBS receives data from administrative agencies in a highly aggregated form, and the coverage of these source data cannot be assessed. For example, the national accounts compilers have no insight on the coverage of the monetary statistics data nor of the balance of payments.

3.3 Statistical techniques

3.3.1 Data compilation employs sound statistical techniques to deal with data sources

As mentioned in 3.2.1, the national accounts staff cannot assess thoroughly the accuracy of the administrative data, because these data are available in highly aggregated format.

The BBS Central Data Processing Unit designs data validation software (for checking coding, editing, and tabulation errors) for each survey questionnaire. The data are further validated for processing errors. Imputations and the adjustments are made for nonresponse and missing observations.

3.3.2 Other statistical procedures (e.g., data adjustments and transformations, and statistical analysis) employ sound statistical techniques

The BBS has not yet developed special techniques for data adjustments for the nonobserved activities. Procedures to obtain exhaustive measures of GDP are related mainly to the improved coverage of the unincorporated sector.

Production approach procedures

Estimates of gross output and intermediate consumption are compiled for 15 major activities and 17 subactivities. The common approach is to extrapolate the benchmark output data, using volume indices from the business statistics or administrative data. Intermediate consumption and consumption of fixed capital are derived using fixed ratios. Almost all ratios are derived from benchmarks that are more than five years old, which deviates from the recommended good compilation practice. Further, the estimation methods for almost all activities assume a fixed relationship between intermediate consumption and output for both the current and constant price estimates.

The table below summarizes the techniques used to derive gross value added for some of the main activities:

Table 2.

Main Techniques Used to Derive Gross Value Added

article image

The following observations are on the techniques used to address some specific compilation issues:

  • The output for owner-occupied dwellings is valued as the estimated rentals that tenants would pay for similar accommodation. This practice is in line with international recommendations; however, the benchmark data used are outdated (the Population Census of 1991), and only two types of locations (urban and rural) are distinguished.

  • Growing crops, standing timber, and stocks of fish are not treated as work-in-progress. The total output from these activities is allocated to the period when the harvest occurs.

  • Data on inventories are not available. No adjustments are made in the national accounts estimates.

  • The consumption of fixed capital is a fixed ratio from the output. The ratio is outdated (from the 1996/97).

  • Data for government transactions are on a cash basis. No adjustments are made to convert government transactions to an accruals basis.

Regarding the compilation of volume measures of GDP, the following are noted:

  • The base year for the constant price estimates is 1995/96, while the international best practice recommends that chain linking be used to derive the constant price estimates. Otherwise, the base year for the estimates should be updated with a minimum frequency of five years.

  • The deflators used to derive the constant price estimates do not measure the underlying price changes of the aggregates. The WPI may include price changes related to transportation and trade margins, which are not part of the value of the output. In addition, the IEPI are compiled for heterogeneous groups, and may not measure pure price changes.

  • Double deflation techniques are not used. In a few cases, volume measures of value added are derived by extrapolation, and current price estimates are compiled through reflation.

  • Single deflation of value added is often used, while the 1993 SNArecommends double deflation or extrapolation methods.

  • Volume measures of import duties are estimated by deflating the current estimates using the implicit deflator of the gross value added. The 1993 SNArecommends that volume measures of taxes and subsidies on products be estimated by applying base year tax rates to the volume of transactions subject to the tax or subsidy. Alternatively, the volume measure could be obtained by extrapolating the base-year tax/subsidy using a volume extrapolator of transactions subject to the tax/subsidy.

Expenditure approach procedures

The estimate of household final consumption expenditure is derived using commodity flow techniques (data from the HIES, business statistics, administrative data). The constant price estimates are derived by deflation using the total CPI. Government final consumption expenditure is estimated for 14 functional categories. Constant price estimates are derived by deflation with the CPI for the middle income employees. The data on gross fixed capital formation cover mainly the government sector. Changes on inventories are not estimated. Expenditures on items that are considered stores of wealth (jewelry, antiques, etc.) are included in the estimates of valuables only if they are purchased by enterprises. Such expenditures by households are included in the final consumption expenditures. Imports and exports of goods and services are deflated using the IEPI, which are compiled from the customs data for goods only.

3.4 Assessment and validation of intermediate data and statistical outputs

3.4.1 Intermediate results are validated against other information where applicable

Intermediate data on major activities, mainly industry, are assessed against related indicators, such as volume and price trends. Data on production of primary goods are assessed against merchandise trade statistics and other secondary sources, including the HIES.

3.4.2 Statistical discrepancies in intermediate data are assessed and investigated

Assessments of potential discrepancies of major intermediate data are limited, owing to the lack of data. As already mentioned, for most of the activities the gross value added is compiled using fixed structural ratios. The validity of the intermediate consumption/gross output, gross value added/gross output, and consumption of fixed capital/gross output ratios cannot be checked.

3.4.3 Statistical discrepancies and other potential indicators of problems in statistical outputs are investigated

No supply and use framework is used to investigate discrepancies to make the GDP by activity and expenditure consistent. The remaining discrepancies between GDP by productive activities and GDP by expenditure components are shown explicitly. The discrepancy is about 1—3 percent.

3.5 Revision studies

3.5.1 Studies and analyses of revisions are carried out routinely and used internally to inform statistical processes (see also 4.3.3)

The direction and magnitude of revisions between preliminary and final data are investigated. The sources of errors, omissions, and fluctuations in the data are also investigated. The results of current revisions are used to inform compilation processes for the subsequent periods. The National Accounting Wing of the BBS informs the users about the results of the revisions between preliminary and final data.

4. Serviceability

4.1 Periodicity and timeliness

4.1.1 Periodicity follows dissemination standards

The dissemination of annual GDP statistics meets the recommendations of the GDDS.

4.1.2 Timeliness follows dissemination standards

The preliminary GDP estimates are published about one month before the end of the fiscal year. These estimates are based on nine months’ observed data and three months’ projections. The final annual GDP data based on 12 months of observations are disseminated with a lag of 11 months, which does not meet the GDDS recommendations of 6–9 months.

4.2 Consistency

4.2.1 Statistics are consistent within the dataset

The GDP by activity and by category has a statistical discrepancy of 1–3 percent.

4.2.2 Statistics are consistent or reconcilable over a reasonable period of time

Consistent annual national accounts time series are available from 1980/81. The latest major revision of the estimates (i.e., the adoption of the new base year 1995/96) has been worked back to 1980/81.

4.2.3 Statistics are consistent or reconcilable with those obtained through other data sources and/or statistical frameworks

The national accounts statistics are fully consistent with the balance of payments and the monetary statistics data. As mentioned in 2.3.1, adjustments are made to the central budget data to be in line with the institutional classification in 1993 SNA.

4.3 Revision policy and practice

4.3.1 Revisions follow a regular and transparent schedule

The revision process follows a regular, well-established schedule. The preliminary national accounts estimates are disseminated one month before the end of the financial year. The final national accounts estimates are released 11 months after the end of the period.

The major revisions for rebasing of the national accounts used to follow a cycle of 10 years. The last one was in 1995/96. Detailed explanation on the changes introduced on the occasion of the new base year 1995/96 has been provided in a special publication published by the BBS in both Bengali and English. The national accounts compilers hope to have a five-year cycle for future rebasing.

4.3.2 Preliminary and/or revised data are clearly identified

Preliminary data are clearly identified as such in the hardcopy publications produced by the BBS. The revised data are disseminated with the same level of detail as those previously published.

4.3.3 Studies and analyses of revisions are made public (see also 3.5.1)

The revisions of the series for the new base year of 1995/96 were widely discussed with users of the statistics even before their publication. Explanations of these revisions are contained in a dedicated publication. Major revisions of data owing to changes in methodology are measured and explained in the media when the revised data are disseminated.

5. Accessibility

5.1 Data accessibility

5.1.1 Statistics are presented in a way that facilitates proper interpretation and meaningful comparisons (layout and clarity of text, tables, and charts)

The national accounts data are available only in hardcopy publication—the BBS National Accounts Statistics. The data are presented clearly, and they are always accompanied by an analysis of the most recent developments. Several tables contain ratios such as shares in GDP or GDP per capita. The interpretation of the data is facilitated by charts. The data sets are published with various levels of detail. Time series from 1980/81 are also available in a hardcopy publication.

5.1.2 Dissemination media and format are adequate

National accounts data are disseminated only through a hardcopy publication. No information releases are prepared for the media. The website of the BBS does not disseminate national accounts data. Longer time series can be accessed only through print publications.

5.1.3 Statistics are released on a preannounced schedule

No predetermined date exists for the release of the national accounts. The BBS officials have no authority to release data to the public without the prior approval of the Minister of Planning. Thus, long lags exist between the completion of the compilation process and the final release of data. This current procedure creates doubts among users about the transparency of the BBS. The BBS has no advance release calendar for any of its data sets.

5.1.4 Statistics are made available to all users at the same time

The users are not informed when the national accounts are to be released. As the release is only in hardcopy publication, it is difficult to ensure the simultaneous release of data to all users. Users of the national accounts data generally perceive that the current dissemination practice is not transparent.

5.1.5 Statistics not routinely disseminated are made available upon request

Unpublished, but nonconfidential, data such as certain breakdowns by regions can be obtained for free. The availability of additional information is not widely publicized by the BBS.

5.2 Metadata accessibility

5.2.1 Documentation on concepts, scope, classifications, basis of recording, data sources, and statistical techniques is available, and differences from internationally accepted standards, guidelines, or good practices are annotated

Information on the data sources and methods used to compile the estimates is provided in the annual national accounts publication. Metadata are also available on the DSBB, but it has not been updated since August 2002. Extensive documentation on concepts and methods is published on the occasion of base year changes, but these publications may be difficult to find after a number of years. The metadata do not provide sufficient information on the deviations from international guidelines or the weaknesses in the source data. Also, they do not describe the possible biases in the source data, the degree of coverage, or the level of nonresponse.

Details about the concepts, definitions, sampling procedures, and other attributes of the surveys and censuses, which are data sources for the national accounts, are provided in the reports of these surveys.

5.2.2 Levels of detail are adapted to the needs of the intended audience

The metadata documented by the BBS provide a general description of the methodology, which may not have the level of details needed for specialized users.

5.3 Assistance to users

5.3.1 Contact points for each subject field are publicized

The publications clearly mention the National Accounting Wing of the BBS as the unit responsible for the compilation of national accounts. Detail information (such as contact point and telephone numbers) is available to assist the users.

5.3.2 Catalogs of publications, documents, and other services, including information on any charges, are widely available

A list of publications is given in each BBS publication. There is information on how they may be obtained.

Table 1.

Bangladesh: Data Quality Assessment Framework (July 2003): Summary of Results for National Accounts (Compiling Agency: Bangladesh Bureau of Statistics)

article image
article image

II. Government Finance Statistics

0. Prerequisites of quality

0.1 Legal and institutional environment

0.1.1 The responsibility for collecting, processing, and disseminating the statistics is clearly specified

The obligation to collect, compile and disseminate fiscal data in Bangladesh has been authorized under three separate legal provisions. Article 127 of the Bangladesh Constitution established the office of Comptroller and Auditor General (C&AG) for statutory audits. This office was assigned responsibility under the Additional Functions Act, 1974 to prepare and submit fiscal data as needed for government budgetary purposes. An amendment to this act later relieved the C&AG from the responsibility of preparing fiscal data and reassigned it to the Controller General of Accounts (CGA), leaving the C&AG its original authority as an independent auditing agency for fiscal data prior to their submission to Parliament. More recently, the administrative responsibility for work on fiscal statistics, particularly subannual data, was specifically delegated to the CGA as authorized by the Rules of Business, 1996, which was further amended in August 2000.

Provisional monthly and annual data are compiled by the CGA. Monthly data are disseminated on the Ministry of Finance (MOF) website, http://www.gobfmance.org, and annual data are submitted to Parliament for approval and dissemination after the final accounts are certified by the Auditor General. A unit in the CGA is responsible for restructuring the annual data, which are prepared in accordance with the national methodology, to the government finance statistics (GFS) format.

0.1.2 Data sharing and coordination among data-producing agencies are adequate

The MOF has operational control of most government transactions and the CGA enjoys good cooperation with the responsible line ministries and agencies. Legal and administrative orders exist to ensure compliance from outside agencies, most notably the Bangladesh Bank Order, 1972, as amended, which directs the Bangladesh Bank to provide data to the MOF. Two notable obstacles to the preparation of complete central government accounts are the continuing (but recently improved) practice of maintaining large suspense accounts, which impede the preparation of data classified by economic use and function, and problems in recording all externally-funded development outlays under the Annual Development Budget (ADB).

Until recently, the MOF had made no systematic attempt to compile comprehensive debt statistics, but instead relied on a variety of sources for these records, including the Bangladesh Bank (BB), the National Savings Directorate, and the External Relations Division at the MOF. Recently, an Executive Order was issued to establish a Resource and Debt Management Wing within the Finance Division, which will be responsible for establishing a comprehensive database for debt statistics.

The government has established an interagency committee which will, inter alia, promote the quality of fiscal data. The Poverty Reduction and Growth Facility (PRGF) Monitoring Committee was formed to monitor economic developments under the IMF program, but an important element of that process is its establishment of a forum to address data issues between agencies. It meets on a monthly basis, chaired by the Joint Secretary of Budget at the MOF, and includes representatives of the National Board of Revenues, the National Savings Directorate and External Relations Division of the MOF, BB, and the Bangladesh Bureau of Statistics.

0.1.3 Individual reporters’ data are to be kept confidential and used for statistical purposes only

Only budget sector accounting and administrative data are used in compiling the national presentation of the budget accounts, which are the basis for preparing the GFS and confidentiality issues do not normally arise. MOF employees are subject to the “Official Secrets Act,” which prohibits any government official from divulging any information of a sensitive nature of which he is in possession. There is no other legislation nor regulation which specifically requires the confidentiality of data collected by the MOF, but all public sector employees including those at the MOF are responsible for obeying relevant civil service regulations establishing their code of conduct, and there have been no experiences in the past where confidentiality has been an issue.

0.1.4 Statistical reporting is ensured through legal mandate and/or measures to encourage response

The line ministries and other government agencies are obligated under the Rules of Business, 1996, as amended in 2000, to provide fiscal data as mandated by the MOF. Compliance by outside government units is normally mandated by other laws and administrative orders, such as the Bangladesh Bank Order, 1972, as amended, and the MOF enjoys the willing cooperation of other government agencies on an informal basis.

0.2 Resources

0.2.1 Staff, facilities, computing resources, and financing are commensurate with statistical programs

A small office (consisting of four CGA accountants and five information technology personnel) is responsible for compiling the annual fiscal data according to the classification system in the Government Finance Statistics Manual 2001 (GFSM 2001), which is accomplished along with their regular tasks. This level of personnel is adequate for current needs, but further progress in implementing the GFSM 2001 methodology or increasing the frequency of reporting will require additional manpower. Moreover, there is a need for more extensive statistical training of staff, which presently consists mainly of on-the-job training, with emphasis on learning processing procedures. Only one staff member, the head of this branch, has attended an IMF course in GFSM 2001 methodology. MOF staff retention is not an issue, as low salaries are offset by a benefits package and job security that the private sector does not provide.

The absence of computing facilities and corresponding software were formerly a major impediment to the efficient compilation and processing of fiscal data, but several donor programs have significantly upgraded the CGA’s processing capacity in recent years. The most significant of these is a joint program of the U.K. Department for International Development (DFID) and the Royal Netherlands Government to completely computerize CGA operations, which includes the provision of computers and necessary software for all processing centers. In general, the provision of other physical resources such as buildings, transport, and other support facilities is adequate.

0.2.2 Measures to ensure efficient use of resources are implemented

The allocation of resources is part of the overall managerial oversight at the MOF. The Minister of Finance and his advisors rank statistical requirements as a high priority of the MOF. Several externally funded projects provide synergies in addition to their primary objective of improving the statistical capacity of the MOF, as they put experts familiar with international standards on the ground who can evaluate related statistical methodologies and compilation systems.

0.3 Relevance

0.3.1 The relevance and practical utility of existing statistics in meeting users’ needs are monitored

As part of the budget preparation process, the Minister of Finance routinely meets with public groups including nongovernment organizations (NGOs) and other international agencies to determine their interest in special data issues. There is no other structured and periodic consultation with users at the MOF, but staff do respond on a case-by-case basis to data requests from the public. MOF staff do not routinely attend statistical meetings and seminars organized by international, regional, or professional organizations.

0.4 Other quality management

0.4.1 Processes are in place to focus on quality

Data quality was explicitly recognized as a primary objective of the MOF with the formulation of the PRGF committee, which in addition to monitoring policy implementation provides a forum for government agencies to meet together for the purpose of reconciling data inconsistencies and addressing other statistical issues. In addition, two teams at the CGA, the Task Force on the Quality of Accounts and the Central Quality Assurance Team monitor financial data submissions received by the CGA to identify those which should be reviewed for accuracy and to resolve identified discrepancies. Furthermore, the Auditor General which has independent authority audits the budgetary final accounts and certifies their accuracy.

The MOF also holds a weekly seminar series which encompass a variety of topics including statistical issues to upgrade staff skills and their awareness of analytical issues. The MOF has no procedure in place to focus on data timeliness-quality tradeoffs.

0.4.2 Processes are in place to monitor the quality of the statistical program

The Parliament reviews the final budgetary accounts prepared in the national presentation and has final jurisdiction over the accounting system applied. The structure of budgetary statistics is reviewed on an ad hoc basis to ascertain if modifications are required. As part of this process, the IMF has provided technical assistance including a Multisector Mission in 2002 which inter alia reviewed the compilation of fiscal statistics and tendered its recommendations for further improvements.

0.4.3 Processes are in place to deal with quality considerations in planning the statistical program

All fiscal data are vetted for accuracy by the Task Force on the Quality of Accounts and the Central Quality Assurance Team at the CGA. The monthly fiscal data posted on the MOF website are not subsequently revised to improve accuracy, but annual data are further processed by the CGA and also audited by the independent office of the Auditor General, which may suggest procedural changes. The MOF responds on a case-by-case basis to address data requirements voiced by the public. Systemic changes are rare and done on an ad hoc basis.

1. Assurances of integrity

1.1 Professionalism

1.1.1 Statistics are produced on an impartial basis

As authorized under the C&AG Act, 1974, and the Rules of Business, 1996, fiscal statistics are compiled from government transactions as recorded by government fiduciary agents throughout the country and collected by the CGA, and as such represent actual transactions rather than estimates. No laws specifically establish the professional independence of staff responsible for producing and reporting fiscal statistics, but there is a longstanding tradition of MOF independence in their dissemination. Typically, the Minister of Finance and his top advisors have academic credentials, extensive prior experience in their professional career, and have established a reputation which promotes data integrity.

The accounting effort has in recent years been supported by substantial external assistance which has automated most of the actual compilation effort and led to reforms in the review and assessment of these data, most notably under the “Financial Management Reform Program” and the “Reform in Budgeting and Expenditure Control Project.” These initiatives have typically been organized under the joint direction of MOF staff and foreign consultants with the objective of instilling international standards in MOF work.

To deepen professionalism in MOF activities and its staff, the MOF has established a practice of holding weekly seminars presented by staff and outside participants, including attendees from NGOs, multinational agencies, and the staff of foreign governments, on a variety of topics including statistical practices to broaden staff expertise.

1.1.2 Choices of sources and statistical techniques as well as decisions about dissemination are informed solely by statistical considerations

Fiscal data are compiled based on actual transactions and do not require sampling, and there is no evidence of political interference in the compilation of fiscal statistics. The administrative sources used in the compilation of GFS are obvious and appropriate, and largely follow IMF recommendations. The authorities have made a commitment and agreed to a migration plan for full adoption of the GFSM 2001 methodology over the longer term, which will require a substantial investment of human and computer resources. One step in this process is a World Bank program which will identify existing differences between the use of cash accounting in Bangladesh and its international application.

1.1.3 The appropriate statistical entity is entitled to comment on erroneous interpretation and misuse of statistics

The Minister of Finance’s presentation of the Budget Speech to Parliament is the keystone of its public awareness program, which is supplemented through the year by the Minister fielding questions from Parliament and holding press briefings on fiscal developments of interest to the public. On those occasions when the press misreports fiscal developments or reports incorrect fiscal data, MOF staff, often led by the Permanent Secretary, directly communicate with the source to correct their error.

1.2 Transparency

1.2.1 The terms and conditions under which statistics are collected, processed, and disseminated are available to the public

The principal sources of fiscal data disseminated by the MOF are the Budget Presentation to Parliament, the annual Bangladesh Economic Review, and the MOF website, each of which identifies its data sources. The website provides some additional information in this area. The collection of data is governed by public laws, which are generally available. General Data Dissemination System (GDDS) fiscal metadata for Bangladesh are also posted on the IMF’s Dissemination Standards Bulletin Board (DSBB), which can be accessed by hyperlink from the MOF website.

1.2.2 Internal governmental access to statistics prior to their release is publicly identified

The Minister of Finance is provided with fiscal data prior to its release to prepare for questions that may arise when the data are made public, and may ask for clarification of specific developments, but does not change or prevent data dissemination. This access is not publicly divulged, but is reported in the GDDS metadata posted on the IMF’s DSBB, and available by hyperlink from the MOF website.

1.2.3 Products of statistical agencies/units are clearly identified as such

All fiscal data released to the public by the MOF are clearly identified as such either through the MOF logo on communications or reference in specific tables.

1.2.4 Advance notice is given of major changes in methodology, source data, and statistical techniques

The methodology used to compile fiscal data is rarely changed, and the MOF does not provide advance notice of these changes. At the time such changes are introduced, it provides the user with an explanation in the form of footnotes to the relevant data.

1.3 Ethical standards

1.3.1 Guidelines for staff behavior are in place and are well known to the staff

All MOF staff are subject to the code of conduct for civil servants as established in the Government Servants Conduct Rules, 1979, which emphasizes the need to prevent conflicts of interest and ensure personal integrity and discipline. There are no codified rules specifically governing the behavior of staff at the MOF which would place additional conditions on their tenure at the Ministry, but since its creation the MOF has enjoyed a strong culture of ethical standards that discourages political interference.

2. Methodological soundness

2.1 Concepts and definitions

2.1.1 The overall structure in terms of concepts and definitions follows internationally accepted standards, guidelines, or good practices

The compilation of both monthly and annual fiscal statistics for the central government largely follows the methodological framework prescribed in A Manual on Government Finance Statistics, 1986 (GFSM 1986).

The government has recently committed itself to adopt the methodology of GFSM 2001 and agreed upon a migration plan with the IMF to implement this initiative over the longer term. As a first step in this process, last year the government submitted data using the GFSM 2001 classification structure for publication in the Bangladesh country page in the IMF’s Government Finance Statistics Yearbook 2004.

2.2 Scope

2.2.1 The scope is broadly consistent with internationally accepted standards, guidelines, or good practices

Both the GFSM 1986 and GFSM 2001 employ the general government sector as their conceptual basis for maintaining accounts. Data are not available in Bangladesh to construct a consolidated set of general government accounts, as despite its adoption of a unitary form of government, comprehensive data on local government operations are not provided to the MOF. Transfers from the central government to the local governments are included in budgetary statistics, but some local governments have revenue sources which are unreported and their overall operations are not compiled. In an operational sense, however, local government operations are relatively insignificant as they have no independent authority to borrow. The budgetary central government accounts in Bangladesh therefore provide a close approximation of general government operations, as there is no social security fund outside the government pension program, the operations of which are included in the budget accounts, and there is only one extrabudgetary account, for the maintenance of bridges, which is relatively small.

Central government data are sufficiently detailed to allow the classification of revenue and expenditure items to the level specified in both GFSM 1986 and GFSM 2001 However, some expenditure items remain in suspense accounts and cannot be further classified. Financing data are classified by primary counterparty institutional sectors, with further detail on the type of financing instruments employed and their term structure.

Debt data are not compiled in the context of GFS reporting, but provided separately by the responsible agencies. The level of detail for these data includes the level by main financing instrument, term structures, and the interest rates carried on foreign debentures.

2.3 Classification/sectorization

2.3.1 Classification/sectorization systems used are broadly consistent with internationally accepted standards, guidelines, or good practices

There are weaknesses in the classifiction and sectorization in the national presentation of fiscal statistics. Local governments are only partially accounted for, with own-revenues and expenditures not recorded. Three nonfinancial public enterprises, the railway, the post office, and the telephone and telegraph, are treated as units within government ministries and data for these entities are incorporated in budgetary accounts. Although these data are for net rather than gross operations, they should be excluded from GFS accounts. The budgetary data also include transfers of both a current and capital nature for these enterprises, which would more correctly be classified as subsidies and grants, respectively. Other government enterprises such as the university system provide goods and services to the public at prices that do not fully cover all costs and receive budgetary transfers in partial compensation. Many of these transfers are recorded in the budgetary accounts as subsidies, in accordance with the methodology of GFSM 1986/2001 and the quasi-fiscal nature of these outlays is appropriately identified, but in other cases as for Bangladesh Jute Mills Corporation the transfers are in the form of net lending, which is unlikely to ever be repaid and hence escapes a full accounting of government outlays on behalf of the private sector.

Revenue, expenditure, net lending, and financing are classified in the national presentation in broad conformity with the GFSM 1986 methodology. More specifically, the classification is as follows:

Revenue and Grants: The presentation of revenue and grants is in close conformity with GFSM 1986/2001 methodology, with government receipts classified by three major headings: tax, nontax revenue, and foreign grants. Nontax revenues include receipts from the three nonfinancial public enterprises cited before (the railways, post offices, and the telephone and telegraph), and grants are underreported due to the incomplete recording of donor assistance in support of the ADB. In accordance with GFSM 1986 but not GFSM 2001 methodology, receipts from privatization are included above the line in nontax revenue, but these receipts are small in any case.

Tax revenue is further identified as deriving from the National Board of Revenue and those from other sources, the former including taxes on (i) income and profit, (ii) value added, (iii) imports, (iv) excises, and (v) the supplementary duty; while other tax income comprises (i) narcotics and liquor duties, (ii) taxes on vehicles, (iii) land revenue, and (iv) the stamp duty. Nontax revenue is identified from a variety of sources, including (i) dividend and profit, (ii) interest, (iii) administrative fees and charges, and (iv) charges for services provided, as well as receipts from the three nonfinancial public enterprises cited earlier. Grants are further identified by principal source, including food, commodity, and project aid, as well as technical assistance.

Expenditures and Net Lending: Expenditures are broadly classified by two major types, current and capital, which are consolidated from the records of line Ministries and other government agencies. Several aspects of the accounting system impair the accuracy of both the aggregate level of expenditure and its composition. Some externally-funded project aid is underreported, leading to an understatement of government outlays, which does not however affect the overall budgetary balance, because as stated above, grants are correspondingly understated. The further classification of current expenditures is impaired by the practice of maintaining large (but reduced in recent years) suspense accounts, and some expenditure under the Annual Development Budget is not sufficiently detailed to permit full classification. Externally funded projects in the ADB are significantly understated due to incomplete reporting of donor agency contributions. Donor agencies are increasingly aware of this systemic problem and have recently convened a donor community Good Practices meeting in Rome to recommend standardized classifications that should improve future budget reporting in aid recipient countries.

Expenditures are identified by nondevelopment revenue and capital outlays, structural adjustment payments, the Annual Development Program, and several smaller accounts, one consisting solely of outlays with respect to Food Account Operations, which is a targeted subsidy to ensure national food sufficiency. The economic classification of the nondevelopment revenue budget provides data on pay and allowances, purchases of goods and services, subsidies and current transfers (other than the Food Account), and interest payments, further classified as domestic and foreign. In short, the classification of expenditure falls short of GFSM 1986 methodology, due to an incomplete classification of expenditure by economic and functional uses.

Net lending is generally classified appropriately, although as mentioned above some loans will be subsequently forgiven and would be more appropriately classified as grants. Loans and repayments are presented as gross flows, separated into financial and nonfinancial institutions and other, and netted out in the summary accounts.

Financing: Financing is categorized as bank and nonbank resources (net), and foreign receipts and repayments. The domestic banking data are further classified as long-term and short-term, and the items identified under National Savings Schemes constitute nonbank domestic borrowing. Foreign loans are distinguished as project aid, non-ADB project aid, special development, and policy loans.

Government Debt: Data on government debt are not disseminated with budgetary data. Domestic debt are disseminated in summary form on the BB website, with T-bills and bonds further distinguished by term structure, and Certificates issued by the National Savings Directorate, which comprise domestic nonbank borrowing, only available as an aggregate. External debt is disseminated in the Economic Relations Division of the MOF annual publication Flow of External Resources into Bangladesh, which in accordance with international standards, classifies debt by currency, type of debt instrument, and type of debt holder. While the detail of domestic debt could be improved, overall the structure of Bangladesh central government debt statistics meet GDDS recommendations. Government-guaranteed debt statistics are not compiled.

2.4 Basis for recording

2.4.1 Market prices are used to value flows and stocks

The prices used to value transactions reflect actual or imputed cash payments, in accordance with GFSM 1986. However, also in accordance with GFSM 1986, the value of debt liabilities are recorded at face value, the amount at which they ultimately will be redeemed, rather than current market value. External debt is valued in domestic currency in accordance with GFSM 1986 guidelines, with foreign debt converted to domestic currency terms using the mid-point exchange rate at the end of the reference period.

2.4.2 Recording is done on an accrual basis

Central government budgetary transactions are effectively recorded on a cash basis. Accounts are maintained on the basis of MOF documentation, which authorizes the transfer of resources to or from the consolidated fund managed by BB on behalf of the MOF. As such, these records are a close proxy for cash receipts and payments, and for practical purposes, the recording is done on a cash basis, in accordance with GFSM 1986. Similarly, financing data are recorded on the basis of administrative information on the incurrence or redemption of liabilities and BB information on changes on the holding of cash in government accounts, which is effectively a cash basis.

While these transactions closely approximate a cash based accounting system, there are small differences in the timing between above-the-line and financing (below-the-line) data, which in principle introduces a small statistical discrepancy into the budgetary accounts. To adjust for this difference, a small statistical entry is introduced into the accounts to enforce equality, but in Bangladesh, this entry is somewhat larger and more volatile than this factor alone would account for. A more important factor is the failure to include some ADB expenditure for which financing has been entered separately. In regard to the BB data on government recourse to the banking system, in some cases there is a substantial delay between the MOF authorization and the actual transfer of funds from the BB, and the MOF data on domestic bank borrowing would not correspond exactly to the BB records for these transactions (the check float issue).

The government of Bangladesh has committed itself to adopting accrual accounting in accordance with GFSM 2001 methodology, but under the agreed migration path Bangladesh will make this transition only in the long term.

2.4.3 Grossing/netting procedures are broadly consistent with internationally accepted standards, guidelines, or good practices

Government finance transactions are shown on a gross basis in the national presentation, in accordance with international standards. Moreover, tax refunds or other adjustments to revenue receipts are correctly classified as negative revenue transactions, rather than an expenditure item. However, the revenue records do depart from international standards in recording these outflows in the accounting period they are effected, rather than offsetting revenues in the period the overpayment actually occurred. These amounts are usually small, and the departure from international standards in this regard is not significant.

3. Accuracy and reliability

3.1 Source data

3.1.1 Source data are obtained from comprehensive data collection programs that take into account country-specific conditions

The accounting system at the CGA provides timely budget data on revenue and expenditures and net lending on a monthly basis. These data are compiled using the chart of accounts to produce a set of fiscal tables to verify that government transactions are in accordance with legal authorization and to provide an analytical basis for fiscal analysis. At the end of the fiscal year, annual data are transformed using the set of accounts to produce GFS statistics for publication in the IMF’s GFS Yearbook.

Financing data are currently assembled from a variety of sources. The BB maintains records on domestic bank borrowing and the issuance of T-bills and T-bonds, the National Savings Directorate collects data on domestic nonbank borrowing, and the Economic Relations Division in the MOF monitors data on external financing. These disparate functions are in the process of being consolidated in the new Resource and Data Management Wing being organized at the MOF. By making greater use of the electronic data processing facilities at the MOF it is anticipated that these data will be available in the future on a more comprehensive and timely basis.

3.1.2 Source data reasonably approximate the definitions, scope, classifications, valuation, and time of recording required

The central government budget classifications for revenue and expenditure—the main source data—broadly align with GFSM 1986 definitions, classifications, time of recording, and valuation, as do the source data for financing operations. GFS aggregates are not automatically derived from accounting data—never in the case of monthly data—but derived from annual data for publication in the IMF’s GFS Yearbook. As software has been developed for this purpose, GFS data are easily reconcilable with the national presentation of budgetary accounts.

3.1.3 Source data are timely

Central government budgetary data are timely reported, owing to computerization of CGA operations. Monthly fiscal data are normally available within 30 days of the end of the reference month, and in principle posted on the website within 60 days, as at present, but in the past, there have been substantial lags. Annual data are audited more carefully, including certification by the Auditor General, and the complete final accounts are generally released by the MOF to other government agencies, but not disseminated, within six months of the end of the reference year. Currently, the delay in disseminating these data is approaching two years.

3.2 Assessment of source data

3.2.1 Source data—including censuses, sample surveys and administrative records—are routinely assessed, e.g., for coverage, sample error, response error, and nonsampling error; the results of the assessments are monitored and made available to guide statistical processes

The national presentation of fiscal statistics on which the GFS is based is directly compiled from accounting and administrative data, and does not employ statistical techniques in the compilation procedure. In regard to the accounting and administrative data, information on fiscal transactions from two different sources can on occasion differ, and the CGA has primary responsibility for resolving these discrepancies, which are later reviewed in the MOF for accuracy. The “reconciliation” process which is the primary means of validating data involves vetting the accounting data provided by line agencies under the supervision of the CGA and benchmarking preliminary data to internally audited data for all budgetary agencies. This process may take some time to complete, and the trade-off between accuracy and availability is resolved in favor of expediency in the release of monthly data, which are not revised once they have been posted on the website. Annual data are thoroughly vetted prior to their release from the MOF, and further reviewed for accuracy by the Auditor General, who has independent authority to assess fiscal data.

3.3 Statistical techniques

3.3.1 Data compilation employs sound statistical techniques to deal with data sources

The national presentation of fiscal statistics on which the GFS is based is directly compiled from accounting and administrative data, and does not require statistical techniques in the compilation procedure.

3.3.2 Other statistical procedures (e.g., data adjustments and transformations, and statistical analysis) employ sound statistical techniques

The compilation of fiscal statistics does not rely on other statistical techniques.

3.4 Assessment and validation of intermediate data and statistical outputs

3.4.1 Intermediate results are validated against other information where applicable

Differences in data between government agencies are addressed in the context of the PRGF Monitoring Committee, which brings together representatives from the concerned agencies on a monthly basis. An item of particular interest is government domestic bank borrowing, information on which is available from both the MOF and BB. The differences between these data are reconciled so far as possible, addressing known conceptual issues such as the check float at BB.

3.4.2 Statistical discrepancies in intermediate data are assessed and investigated

Two units of the CGA focus on data consistency and quality issues, which are primarily addressed in the context of preparing final annual data, as monthly data are not revised after a preliminary review for obvious discrepancies. The Task Force on the Quality of Accounts has the authority to propose systemic changes to resolve proven problematic areas, while the Quality Assurance Team reviews individual data submissions to resolve any statistical discrepancies that may arise from individual submissions which appear to be anomalous.

3.4.3 Statistical discrepancies and other potential indicators of problems in statistical outputs are investigated

The PRGF committee provides a monthly forum to review possible data discrepancies between government agencies. The principal scope of inquiry is the level of bank borrowing by the government, and the committee compares MOF and banking data on a monthly basis to determine if a reconciliation study is required.

The MOF separately attempts to reconcile the measure of the revenue/expenditure difference from the budgetary accounts with available MOF financing data, but substantial differences remain after this process. It is generally presumed that the financing data are more accurate and the discrepancy arises from incomplete expenditure data.

3.5 Revision studies

3.5.1 Studies and analyses of revisions are carried out routinely and used internally to inform statistical processes (see also 4.3.3)

The MOF does not prepare revised statistics, but does review accounts at the end of the fiscal year to prepare its final accounts. The MOF makes no systematic analyses of the revisions resulting from this practice, but deviations between preliminary and final data are examined to ensure that the reasons for significant differences are understood. In preparing audited accounts, the Auditor General prepares a paper on the corrections he applies to MOF data and the principal reasons for differences between budgetary authorizations and the final budgetary outturn.

4. Serviceability

4.1 Periodicity and timeliness

4.1.1 Periodicity follows dissemination standards

The MOF disseminates on its website monthly budgetary data from its national presentation that includes major central government aggregates, and more complete annual data are compiled at the end of the fiscal year and submitted for parliamentary scrutiny and subsequently disseminated. Annual data in the GFS format are only published in the IMF’s GFS Yearbook and the MOF website. These practices meet the GDDS recommendation for periodicity, which are for the dissemination of major fiscal aggregates quarterly and detailed accounts annually. Bangladesh does not disseminate data for the general government sector, an encouraged extension under the GDDS.

The MOF disseminates central government external debt statistics in its annual publication Flow of External Resources into Bangladesh and domestic debt statistics are disseminated monthly on the BB website. These sources provide, in the case of external debt, details by donor, currency obligation, and types of obligation, and for domestic debt, type of debt instrument, and in the case of government T-bills and bonds, the term of debenture. GDDS periodicity standards are for the annual dissemination of central government debt and quarterly dissemination of major debt categories. Bangladesh nearly meets these recommendations, as domestic debt is disseminated monthly and foreign debt annually—leaving foreign debt aggregates not disseminated quarterly.

Two extensions of these data dissemination standards are strongly encouraged for the government sector under the GDDS, neither of which are observed in Bangladesh. The government does not compile local government fiscal data, and therefore does not disseminate annual general government data, nor does it disseminate government-guaranteed debt on an annual basis.

4.1.2 Timeliness follows dissemination standards

The dissemination of fiscal statistics does not meet GDDS recommendation for timeliness, which are to disseminate central government quarterly aggregates within three months of the end of the reference period, and central government annual operations within one year. The delays for posting monthly MOF data on its website have approached six months, but meet the recommendation at present. Annual, audited data of central government operations are generally compiled and audited within nine months, but are subject to further review, including hearings by the Parliamentary standing committee “Public Accounts Committee” before they are disseminated, and the delay is currently approaching two years.

Comprehensive central government debt dissemination in Bangladesh meets GDDS recommendations. Data on domestic debt is disseminated on the BB website within three months of the end of the reference period, and data on foreign borrowing disseminated within nine months in the MOF publication Flow of External Resources into Bangladesh, both within the recommended nine months. It is envisioned that full implementation of the new Resource and Debt Management Wing at the MOF and its computerization of comprehensive debt records will further reduce the time required for dissemination of these data.

4.2 Consistency

4.2.1 Statistics are consistent within the dataset

There are three principal consistency checks within the fiscal accounts. First, the monthly data in principle should sum to the annual data. Second, the revenue/expenditure or above-the-line balance should agree with that generated by the financing or below-the-line data. Third, the financing data or flow data for the fiscal year should correspond to changes in the outstanding stock of debt over that period. In each case, there are conceptual reasons why the two comparable measures would not be in exact agreement, but in Bangladesh the differences are larger than would be expected.

The monthly fiscal data are vetted for accuracy by the CGA before they are posted on the MOF website, but the primary objective for this series is to obtain a working statistical base for operational purposes, and the quality/availability tradeoff is resolved in favor of timeliness. They are never revised for accuracy once they are posted on the website. Annual data, however, are a final record of fiscal operations for the year, audited by the Auditor General and reviewed by Parliament, and these are thoroughly vetted for accuracy. It is not surprising, therefore, that the monthly data do not sum to the final budgetary outturn.

There are two main conceptual reasons for the revenue/expenditure balance to differ from budgetary financing data. First, above-the-line transactions are recorded in MOF accounts on the basis of budgetary authorizations, usually when checks are issued, which does not coincide with the timing payment, actually is made through the banking system (the before-identified check float issue). Financing data, on the other hand, should closely reflect the actual cash basis for these transactions, and the two would be expected to differ. Second, there can be differences in the timing when budgetary revenues/expenditure transactions are posted in the MOF accounts and when financing instruments are recorded. While the discrepancy between the above-the-line and below-the-line balances is examined and reconciled as far as possible, there has been no systematic analysis of its tendency to cancel out over time (which it would if the differences were merely due to timing), to quantify more significant consistency issues.

Net central government borrowing data over the year is determined by establishing the value of new indebtedness less amortization payments during the year, from both foreign and domestic sources. As the debt agreements establish the actual obligation, these transactions are valued on the basis of their face value rather than market value, if the two differ. Financing data, on the other hand, are based on the actual cash flows associated with amortization of loan contracts, and the two can be expected to differ. There has been no analysis of the relative importance of this factor in reconciling the financing data with the above-the-line deficit in Bangladesh.

4.2.2 Statistics are consistent or reconcilable over a reasonable period of time

In 1998, in conjunction with its introduction of an electronic processing system for budgetary data, the MOF introduced a new classification system which introduced a discontinuity in the fiscal statistics which had been compiled and disseminated using a consistent classification system for the previous decade. To provide users with a suitable bridging period, the MOF disseminated fiscal data compiled by the new classification system for the preceding five years, and undertook a maj or publicity campaign to make users aware of the transition. There has been no systematic analysis to the data to determine their consistency. Standard budgetary measures which compare budgetary aggregates such as revenues, expenditures, and balances to other statistics such as GDP exhibit reasonable stability.

4.2.3 Statistics are consistent or reconcilable with those obtained through other data sources and/or statistical frameworks

The most common comparisons between fiscal statistics and those generated outside the budgetary accounts are for government recourse to bank financing, a measure for which is also generated in the monetary statistics, and for government foreign transactions, which are also recorded in balance of payments data. While there are conceptual reasons why data from two different sources should disagree, primarily differences in the timing these transactions are posted, the differences are larger than can be attributed to this factor.

4.3 Revision policy and practice

4.3.1 Revisions follow a regular and transparent schedule

The MOF does not revise monthly fiscal data, which are posted on the website as preliminary actuals and never revised. After the end of the fiscal year, the MOF keeps the accounts open for a 30-day complementary period to ensure that all transactions attributable to the previous fiscal year are correctly entered. Afterward, the accounts are closed and the CGA prepares a set of final, comprehensive data which constitutes the only revision of actual budgetary transactions. These accounts are submitted to the independent office of the Auditor General, which undertakes its own review of the accounts and issues a report on accounting issues, as well as a comparison of budget submissions as augmented by budget supplements with final data on actual budget execution.

4.3.2 Preliminary and/or revised data are clearly identified

Preliminary data are not prepared outside the monthly data posted on the MOF website, which are so identified. Final/revised data are prepared in the context of the final report of the MOF, and labeled as the final accounts.

4.3.3 Studies and analyses of revisions are made public (see also 3.5.1)

The MOF prepares summary comparisons for inclusion in the Minister’s budget speech, but does not otherwise routinely release any studies of comparisons between budget and actual data. The independent office of the Auditor General subsequently releases, in its report on the final accounts prepared by the MOF, an analysis of budgetary execution which includes identification of any significant data issues and a comparison of budget objectives and their realization.

5. Accessibility

5.1 Data accessibility

5.1.1 Statistics are presented in a way that facilitates proper interpretation and meaningful comparisons (layout and clarity of text, tables, and charts)

Budget documents include the Budget in Brief which presents an overview of budgetary statistics and provides a useful summary of the most important issues, complete with pie charts to illustrate main points. The Budget Speech, given to Parliament and disseminated in written form, gives the Minister of Finance an opportunity to identify main issues and developments. GFS data are disseminated through the IMF’s GFS Yearbook and the MOF website.

5.1.2 Dissemination media and format are adequate

The presentation of fiscal data in the national presentation includes the budget speech and accompanying budgetary documents, the monthly budgetary data on the website, and the annual Bangladesh Economic Review. Additional data are available from the BB and the Auditor General. GFS data are disseminated through the IMF GFS Yearbook and the MOF website. These provide adequate, but not ideal, dissemination media and formats.

5.1.3 Statistics are released on a preannounced schedule

Outside the occasion of the budget presentation to Parliament, there is no preannounced release schedule for the release of fiscal statistics. The monthly fiscal data posted on the website are in principle available within two months of the end of the reference month, as at present, but in practice there have been lengthy delays before actually being posted.

5.1.4 Statistics are made available to all users at the same time

Budgetary data are made available to all users simultaneously, without preferential treatment for selected users. As noted above, the GFS data are not disseminated by the MOF, but are submitted for publication in the IMF’s GFS Yearbook.

5.1.5 Statistics not routinely disseminated are made available upon request

It is not officially publicized, but the MOF does make specialized fiscal data available on a case-by-case basis.

5.2 Metadata accessibility

5.2.1 Documentation on concepts, scope, classifications, basis of recording, data sources, and statistical techniques is available, and differences from internationally accepted standards, guidelines, or good practices are annotated

Detailed documentation of concepts, scope, classifications, basis of recording, data sources is available through publications issued by the MOF. The Chart of Accounts identifies the budgetary codes and their classification, and the manual How to Keep Accounts provides a glossary of budgetary terms and information on the budgetary presentation. Metadata for GFS are only available in summary form on the IMF’s DSBB, which is accessible by hyperlink from the MOF website

5.2.2 Levels of detail are adapted to the needs of the intended audience

Detailed technical descriptions on concepts, sources, and methods are available which allow expert users to assess and interpret the merits of fiscal data. Less detailed information on these factors more directed at general users is also available on the DSBB and, via hyperlink, the MOF website.

5.3 Assistance to users

5.3.1 Contact points for each subject field are publicized

The MOF’s annual publication Bangladesh Economic Review contains an invitation for interested readers to contact the MOF personnel if they have questions or want additional information, and its website contains a contact page for user comments/inquiries. The MOF has a longstanding practice of providing international agencies and donor countries selected data on fiscal developments of special interest. Otherwise, contact information for staff delegated to provide the public with further access is available on the DSBB and the MOF website.

5.3.2 Catalogs of publications, documents, and other services, including information on any charges, are widely available

There is no catalog of publications containing government finance statistics, but government regulations require dissemination of data sources and the public is aware of available sources.

Table 2.

Bangladesh: Data Quality Assessment Framework (July 2003): Summary of Results for Government Finance Statistics

(Compiling Agency: Ministry of Finance)

article image
article image

III. Monetary Statistics

0. Prerequisites of quality

0.1 Legal and institutional environment

0.1.1 The responsibility for collecting, processing, and disseminating the statistics is clearly specified

No legislation, such as a statistics law or a decree, specifically assigns the Bangladesh Bank (BB) the responsibility for compiling and disseminating monetary statistics. However, by tradition the BB has assumed this function, and its role has never been questioned.

Furthermore, several articles in the Bangladesh Bank Order, 19721 deal with the collection and dissemination of data by the BB. Chapter III, Article 36 (3) states that “… every scheduled bank shall send to the Bank [BB] returns … containing such information as may be deemed necessary for carrying out the purposes and objects of this Order at such periods of time as the Bank may from time to time direct.” Article 38 extends this authority to request information from “any Corporation with which it [the BB] has any transactions under Article 16.” The same Order empowers the Research Department and the Statistics Department (SD) of the BB “to request from any natural or juridical person, as well as the Government, any data or information necessary for the proper execution of its duties” (Chapter IV, Article 69). Dissemination of data is covered in Chapter III, Article 39, which states that “the Bank shall compile and publish in such manner and at such times as the Board may consider appropriate a consolidated statement from such information as may be received under this Order.”

The SD, through its General Statistics Division (GSD), has consistently assumed the responsibility for compiling and disseminating monetary statistics. This responsibility is outlined in the BB’s Service Standard 2003, where a description of the tasks to be performed can be found. The Off-site Supervision Department (OSD) and the Banking Regulation and Policy Department also collect data from banks for supervision purposes.

0.1.2 Data sharing and coordination among data-producing agencies are adequate

Different departments of the BB interact during the process of compiling monetary statistics. The GSD receives from banks and other deposit-taking institutions hard copies of their monthly returns (“SBS-1” for commercial banks, and “NBDCs” for other deposit-taking institutions). From the Accounts and Budgeting Department (ABD) of the BB, the GSD receives every week a condensed balance sheet of the BB plus some supplementary information needed for producing the BB’s analytical accounts—also in a hard copy. The hard copies of the banks’ returns are forwarded to the Information Systems Development Department (ISDD) for validation, processing, and the production of the monetary survey. The ISDD supplies the SD with a hard copy of the survey, plus an electronic version for transmission to the IMF Resident Representative Office. The GSD manually processes the data received from the ABD to generate the sectoral balance sheet of the central bank and the central bank survey. Using information provided by the Forex Reserve & Treasury Management Department, the GSD reconciles the holdings of treasury bills and treasury bonds reported by the banks and their foreign exchange reserves. From the OSD, the GSD receives only a weekly report of little or no use for monetary data compilation. Data compiled by the GSD are transmitted to the Monetary Policy Department (MPD), a main user of monetary data.

There is room for improvement in the flow of monetary data between different departments of the BB. Data are not received by the GSD in electronic formats, forcing it to perform a series of manual operations when compiling the surveys. So, the GSD must adjust manually the condensed balance sheet of the BB to produce the central bank survey. Similarly, the GSD does not have electronic access to the information processed by the ISDD, and depends on this department for any further analysis it wants to make using already processed monetary data. Finally, the work of the GSD could be enhanced if it would regularly receive supplementary data from the OSD (e.g., list of largest depositors, list of largest loan exposures), which could be used to verify the quality of the information reported by the banks for monetary statistics purposes.

The SD communicates regularly with other units collecting statistics, namely the Bangladesh Bureau of Statistics (BBS), different public sector corporations, and the stock exchange. The BB receives data from these agencies, which are disseminated through its monthly and quarterly publications, and provides them with monetary statistics. There are also fluid informal contacts between staff of the SD and the other institutions.

0.1.3 Individual reporters’ data are to be kept confidential and used for statistical purposes only

Article 78 (1) of the Order assures the confidentiality of the received data, stating that “… every employee of the Bank shall preserve and aid in preserving secrecy with regard to all matters relating to the affairs of the Bank not published by it, … .” Additionally, Article 78 (2) foresees imprisonment of up to six months and/or fine of up to 500,000 taka,2 in case of violation of data confidentiality, although making an exception if the employee is required by law to disclose the information in the discharge of his duty.

Banks are aware of their rights and obligations regarding the provision of information to the BB. They are also aware that the information they provide will be kept confidential and used only for statistical purposes.

Residual or indirect disclosure of individual data does not occur because the BB aggregates information of 50 institutions. Therefore, the BB has not developed special aggregation rules to protect against these problems. In some cases, and upon request, individual data have been made available for research purposes. In these cases, the higher management of the BB must give prior authorization to the SD to release the data, and the recipient must assure that the data will be kept confidential and used only for research purposes. Released data are not coded, to avoid identification of individual institutions by the user.

Hard copies of the banks’ submissions are kept at the GSD and destroyed after two years. Electronic records are stored by the ISDD in its IBM S400 midstream system, which proved to be safe against external intrusion and is not connected to the Internet. Only authorized persons can enter data in the system, having partial access to them. Every day, data are backed-up and the copies stored in a separate location outside the BB’s main building.

0.1.4 Statistical reporting is ensured through legal mandate and/or measures to encourage response

As mentioned in item 0.1.1, the general framework for reporting monetary data is contained in the Bangladesh Bank Order, 1972. Failure to comply with the provision of timely and substantially correct information carries a fine of up to 25,000 taka3 for each day during which the failure continues (Article 36 (6) and (7) of the Order).

The BB has produced a booklet with detailed guidelines to fill in the return forms. The guidelines provide general instructions, describe the concepts that should be recorded in each item of the returns, and provide a list of the institutions constituting the government sector, the nonfinancial public corporations sector, and the banking sector. The current booklet, issued in May 2001, will shortly be substituted by revised guidelines adapted to the newly introduced returns. Additionally, the BB’s Training Academy offers courses aimed at commercial bank staff responsible for data reporting. Nine of such courses were offered during fiscal year 2004, which were attended by employees of the commercial banks’ headquarters and their branches.

0.2 Resources

0.2.1 Staff, facilities, computing resources, and financing are commensurate with statistical programs

Although SD staff are qualified for the functions they perform, the ABD does not employ a certified accountant. Computer resources are inadequate for efficient data compilation. For data entry and generation of intermediate output tables, the SD has to rely on the ISDD, which uses an outdated program. Accounting records at the BB are kept manually, with all adjustments performed at the end of the fiscal year.

At the beginning of 2005, the SD had 51 staff in its five divisions.4 The GSD is divided into three wings (each under the authority of a Joint Director), one of which (Monetary Statistics Wing (MSW)) is responsible for the compilation of monetary data. A total of seven persons are engaged full-time in the compilation of monetary statistics, supervised by one General Manager and one Deputy Manager. With the current working procedures (e.g., dependence on paper based processing), the GSD seems understaffed, but improvements in the provision of computer hardware and software, as well as efficiency gains in the compilation process, should render the current staffing adequate. All staff responsible for monetary statistics have a post-graduate degree in statistics—a precondition for being recruited at the SD—and the General Manager has also pursued postgraduate studies in England. All have substantial experience in the subject, having worked continuously at the SD from 8 to 25 years. Two staff of the GSD have participated in courses organized by the IMF, although only one participated in the Monetary and Financial Statistics Course. IMF courses are the only foreign training attended by staff of the GSD, but they regularly participate in domestic training and seminars, most of them organized by the BB’s Training Academy. The BB nominates candidates to IMF courses following a seniority criterion, rather than identifying staff who work in areas specifically related to the content of the courses and who will keep working in those areas after the course is completed. Staff of the SD neither participate in monetary statistics courses or seminars offered by other international organizations, nor visit other central banks to learn from their experience.

Different is the human resources situation at the ABD, the department responsible for the BB’s balance sheet. The ABD does not employ any accountant certified by the Institute of Chartered Accountants, Bangladesh. The BB’s Training Academy offers courses in accounting principles and techniques, but personnel of the ABD are not kept up-to-date with the latest developments in international accounting standards.

Salary levels at the BB, covered by the government pay scale, are between one fourth and one half of the salary for similar positions in the banking sector. Turnover is nevertheless low at the SD, in part due to job specialization, job security, and pension benefits accorded to civil servants. Salary increases are not directly related to individual performance, which is assessed once a year. However, staff that systematically receive an outstanding evaluation can advance faster in the career path and even supersede higher ranking employees when promotions are awarded.

Computer resources are inadequate for efficient compilation of monetary statistics. Only the General Manager has his own personal computer with dial-up Internet access. The Deputy General Manager has also one personal computer, but without Internet access, while the seven people working in the MSW have to share three personal computers not linked through a network and not linked to the Internet. The ABD records all transactions manually, with no computerization of general ledgers and sub-ledgers, recording all accruals and other adjustments at the end of the financial year. The ISDD uses an outmoded computer program (COBOL) to generate monetary statistics, although it is in the process of moving to a PC based system and plans to link into a network of the headquarters with all branches of the BB. Physical facilities and other resources seem adequate to perform the required tasks.

The SD has little influence in the budgeting processing decision and almost no flexibility when spending the allocated resources. Resources are being distributed by the Expenditure Management Department (EMD) without reference to a multiyear planning horizon for the development of statistical activities and potential future requirements.

0.2.2 Measures to ensure efficient use of resources are implemented

The process of compilation of monetary statistics at the BB is inefficient, as reflected in mostly paper-based procedures. Data are transmitted in a hard copy by the banks to the GSD, which transfers them (together with handwritten instructions) to the ISDD for the production of the tables used in the monetary survey. Apart from the lack of confidence in the final output that this procedure generates—in spite of verifications performed by the GSD—some rigidities add inefficiencies to the system. For instance, revisions needed because of classification changes cannot be implemented promptly by the GSD, because it must rely on the work of the ISDD to introduce the changes in the computer program.

When producing the central bank survey, the GSD receives hard copies of the BB’s condensed balance sheet and supplementary information provided by the ABD. Neither the ABD nor the GSD conceived the possibility of transmitting the data in an electronic format to facilitate the work of the GSD.

Owing to the lack of computerized applications, GSD staff manually verify data to detect reporting errors or inconsistencies. Based on their experience in the job, staff of the GSD can perform this task thoroughly, albeit in an inefficient manner.

The BB does not have a long-term strategic plan with development objectives for the statistics activities in general and monetary statistics in particular. The allocation of budgetary resources between competing activities is not transparent and decided by the EMD.

0.3 Relevance

0.3.1 The relevance and practical utility of existing statistics in meeting users’ needs are monitored

No formal and structured process of consultation monitors needs of monetary data providers and users. However, the SD receives feedback from commercial banks, the parliament, the Ministry of Finance (MOF), and international organizations, and takes into account these comments and suggestions to improve the compilation and dissemination of monetary statistics. In the recent past, the BB has received technical assistance from the IMF in monetary statistics and has strived to implement the recommendations of these missions.

In its publications, the BB encourages suggestions and comments for improvements in their content, providing an e-mail address where users can direct their queries. However, the BB does not participate regularly in statistics meetings and seminars organized by international and regional organizations. These meetings should provide opportunities for monitoring the relevance of the monetary statistics it produces and the requirements presented by new developments in the financial markets.

0.4 Other quality management

0.4.1 Processes are in place to focus on quality

Staff of the SD, and the GSD in particular, are fully conscious of the importance of quality in building trust among monetary statistics users, and they are sensitive to the different dimensions of data quality. The BB’s Training Academy offers courses on operational excellence, and staff are asked to provide lectures after returning from training programs. This notwithstanding, the BB does not have a formal process to implement externally recognized processes or activities that focus on quality. The BB does not publish any information regarding its commitment to quality.

0.4.2 Processes are in place to monitor the quality of the statistical program

An editorial committee of the BB’s publications regularly reviews the relevance of the data disseminated and their format. This internal exercise focuses on the final product rather than on the process of generating the published data. For external advice on monetary statistics, the BB relies almost exclusively on the technical assistance provided by the IMF.

0.4.3 Processes are in place to deal with quality considerations in planning the statistical program

Owing to the characteristics of monetary statistics (based on accounting records), their accuracy is not highly affected by the timeliness of their dissemination. The GSD is fully aware of the trade-off between timeliness and accuracy and has given priority to timeliness in the dissemination of monetary data. Data are published on a final basis, but the latest available figures are subject to revision the following month if corrections are needed (e.g., if mistakes in the reporting were detected). Very few revisions have occurred in the past.

1. Assurances of integrity

1.1 Professionalism

1.1.1 Statistics are produced on an impartial basis

Although the BB does not have full operational autonomy from the MOF, and the legal environment does not protect BB’s board directors against arbitrary removal, monetary data are compiled following technical criteria.

The BB recruits its personnel in line with its working needs and through an open competitive process that includes exams and interviews. After one year of working at the BB, staff become permanent employees of the bank and are not subject to arbitrary dismissal. This framework ensures their professional independence and protects them against undue pressures or interference by other government agencies.

Staff of the SD receive training at the BB Training Academy, in the form of courses and lectures given by visiting scholars. They also attend courses and seminars offered by other institutions outside the BB, but few have attended international courses. Internal methodological discussions take place, and some staff have produced research papers, some of which have appeared in the BB’s publications.

1.1.2 Choices of sources and statistical techniques as well as decisions about dissemination are informed solely by statistical considerations

The source data for monetary statistics are the balance sheets of all commercial banks and the BB, restricting the possibilities of manipulation through changes in the sample or records used for compilation. Other financial statistics not based on accounting records, such as interest rates, are produced using responses by all institutions or a representative sample. For the classification of financial instruments and sectorization of accounts, the GSD follows the methodology of the Monetary and Financial Statistics Manual (MFSM).

The only exception to a strict adherence to statistical concepts is the classification of the accounts of the Bangladesh Railway (BR), the Bangladesh Telegraph and Telephone Board (BTTB), and the Bangladesh Post Office (BPO). They are classified as accounts of the central government in the BB’s balance sheet rather than as accounts of the public nonfinancial corporations sector.5 The GSD is fully aware of the misclassification in the case of the BR and the BTTB but needs a prior authorization by the MOF to reclassify these accounts. In the case of the BPO, the GSD wrongly considers it an agency of the central government and has not taken any action to reverse its misclassification.

The dissemination of monetary data follows an established procedure with set deadlines that are consistently met, although not announced to the public.

1.1.3 The appropriate statistical entity is entitled to comment on erroneous interpretation and misuse of statistics

In cases of misinterpretation of published data, the Public Relations and Publications Department is responsible for commenting on them through press releases and explanatory letters based on inputs from the concerned departments of the BB.

1.2 Transparency

1.2.1 The terms and conditions under which statistics are collected, processed, and disseminated are available to the public

Bangladesh participates in the IMF’s General Data Dissemination System (GDDS). Its metadata on monetary statistics are accessible on the IMF’s Dissemination Standards Bulletin Board (DSBB), although the country has not updated the posted information since November 2003. The metadata are not posted on the BB’s website.

The Bangladesh Bank Order, 1972, containing the legal terms and conditions for collecting monetary information, is posted on the BB’s website. However, the relevant provisions are scattered throughout the Order and subsumed in sections dealing with other issues, hence hindering easy access by the public to the framework that guides the compilation of monetary statistics.

The BB’s publications, also posted on its website, explain the terms and conditions under which monetary statistics are compiled. They also inform the public about the BB and its products.

1.2.2 Internal governmental access to statistics prior to their release is publicly identified

When the GSD staff complete the monetary tables, they send them to the BB’s Deputy Governors and the Economic Advisor for information purposes. In the meantime, the General Manager of the SD in charge of monetary statistics reviews the data and approves their dissemination. After his approval the tables are sent for publication in Economic Trends (hard copy and on the BB’s website). Data are provided prior to their release to the MPD, the Research Department, the MOF, and the IMF Resident Representative Office, and occasionally to other senior officials. The MFD provides the Governor and the Board of Directors with monetary and economic indicators, which are discussed in a monthly meeting. The public is not made aware of this earlier access.

1.2.3 Products of statistical agencies/units are clearly identified as such

All publications of the BB are clearly identified as products of the SD, with the name and seal of the BB on the cover. All tables contain the source of the data, which are further explained in the notes. The BB does not require attribution when its statistics are reproduced elsewhere.

1.2.4 Advanced notice is given of major changes in methodology, source data, and statistical techniques

Advance notice is not given to the public about major changes in methodology. However, the explanatory notes of the publications indicate and describe methodological revisions.

1.3 Ethical standards

1.3.1 Guidelines for staff behavior are in place and are well known to the staff

The Bangladesh Bank Staff Regulations, 2003 addresses issues of ethical standards in articles covering conflict of interest, acceptance of gifts, foreign awards, conduct towards female colleagues, and declaration of personal property (must be submitted every five years). New staff must sign their acceptance of the Regulations when joining the BB, and receive orientation courses and desk training. No SD staff have been dismissed because of violation of the ethics code.

2. Methodological soundness

2.1 Concepts and definitions

2.1.1 The overall structure in terms of concepts and definitions follows internationally accepted standards, guidelines, or good practices

The methodological framework used by the GSD to compile the central bank survey and the other depository corporations (ODC) survey is consistent with the guidelines of the MFSM. Some previous deviations from these guidelines were resolved when in June 2004 the GSD implemented most of the recommendations of the latest IMF’s monetary and financial statistics mission.

A monetary survey is compiled consolidating the central bank survey with the ODC survey. The main monetary and credit aggregates identified within this framework are reserve money (currency issued plus deposits of banks at the BB), narrow money (currency outside banks plus demand deposits at commercial banks and the BB), M2 broad money (narrow money plus time deposits at commercial banks), M3 broad money (M2 plus deposit liabilities of nonbank depository institutions), net foreign assets, domestic credit (net credit to government plus credit to public nonfinancial corporations and the private sector), and the money multiplier.

2.2 Scope

2.2.1 The scope is broadly consistent with internationally accepted standards, guidelines, or good practices

At end-February 2005, the depository corporations sector of Bangladesh comprised the BB, 49 scheduled banks, the Bangladesh Samabaya Bank (BSB), 74 cooperative banks, three microcredit institutions, 42 land mortgage banks, 23 private finance and leasing companies, the national savings scheme (includes the Postal Savings Scheme), five offshore banks, and cooperative societies/banks under the Registrar of Cooperatives and the Bangladesh Rural Development Board. Except for the cooperative societies, which account for about 2 percent of the deposits of the system, all other deposit taking institutions report on a regular monthly basis to the SD. Data reported cover the headquarters and all branches.

Scheduled banks and the BSB report using the SBS-1 Return, which provides a balance sheet presentation with a breakdown by instrument and economic sector. Additional information is provided in four annex tables with a breakdown of deposits by type and sector, a detail of foreign assets and liabilities by currency, loan classification and provisioning, and the breakdown of investments. Nonbank depository corporations report using the NBDCs Return, which is a simplified version of the SBS-1 Return, and three annex tables, with supplementary information on deposits by type and sector, loan classification and provisioning, and breakdown of investments.

Although the SD receives data from offshore banks, it does not include their accounts in the ODC survey. The BB’s foreign exchange policy considers them as part of the nonresident sector since these banks do not transact with residents.

Recently, following the IMF’s recommendations, the SD has expanded the monetary survey and incorporated the accounts of the nonbank depository institutions in a broader definition of money. The newly disseminated data also include the deposit liabilities of the national savings scheme and its net credit to government, which at end-2004 was one and a half times the net credit to government of the banking sector.

2.3 Classification/sectorization

2.3.1 Classification/sectorization systems used are broadly consistent with internationally accepted standards, guidelines, or good practices

Sectorization of monetary statistics improved substantially after the GSD implemented the recommendations of an IMF monetary and financial statistics mission that visited the country in November 2003. A more detailed sectorization scheme was adopted that divides the domestic sector into central government, local governments, public nonfinancial corporations, other nonfinancial corporations, insurance companies and pension funds, other financial intermediaries, financial auxiliaries, households, and nonprofit institutions serving households. The new returns introduced in June 2004 contain this subdivision. While some banks expressed technical difficulties in splitting the accounts of the resident private nonfinancial sector, the GSD is closely monitoring compliance with the stated guidelines.

One exception to the proper sectorization is the treatment of the BR, the BTTB, and the BPO as units of the central government sector in the balance sheet of the BB, rather than as public enterprises. The SD correctly considers the BR and the BTTB as public enterprises in the guidelines distributed to the banks (which properly classify them in their returns). However, it cannot revise their classification in the BB’s balance sheet until receiving formal approval from the MOF.6 With respect to the BPO, it is considered part of the central government in both the BB’s balance sheet and the banks’ accounts.

Another recommendation of that IMF mission was to provide for a separate identification of advances to nonresidents, which was also incorporated in the revised returns. With this modification, the concept of residency used in Bangladesh’s monetary statistics fully complies with the definition of the Balance of Payments Manual, 5th edition. The concept of residency explicitly acknowledges the situation of Bangladeshi nationals working and living abroad, treating them as nonresidents.

Financial instruments are classified by type, maturity, and currency denomination (local and foreign). The IMF’s mission recommended providing for a separate category for securities issued by commercial banks to investors. The new return has such a category, dividing the holders of bank securities into nonbank public depository corporations, other public corporations, nonbank private depository corporations, other private sectors, deposit money banks, and nonresidents. Repurchase agreements are correctly treated as collateralized loans and are separately identified and sectorized in the new returns. The BB has properly reclassified some nonreserve foreign assets (a deposit in Pakistan subject to dispute and a deposit denominated in a nonconvertible currency) from reserve assets to other foreign assets, while nonmonetary gold is correctly excluded from BB’s international reserves.

To help banks prepare their returns, the BB has issued the Guidelines to Fill In the Banking Statistics Returns SBS-1, 2 & 3. The current Guidelines, from May 2001, are currently being revised to incorporate the changes in the returns of June 2004. The Guidelines provide general instructions on how to fill in the returns and the annex tables and detailed definitions and explanatory notes of the different items contained in the returns. More importantly, the Guidelines provide a comprehensive list of government units, nonfinancial public enterprises, and commercial banks, plus definitions of other financial institutions, minimizing the risk of misclassification by the reporting banks. The Guidelines are supplemented by the regular training that the BB offers to banks’ staff responsible for data reporting.

2.4 Basis for recording

2.4.1 Market prices are used to value flows and stocks

Foreign-currency-denominated financial assets and liabilities are converted into taka using the midpoint market rate. The valuation adjustment takes place at the end of the month. To adjust their foreign-currency-denominated assets and liabilities, banks apply the exchange rate provided by the BB, eliminating valuation discrepancies. BB’s monetary gold holdings are adjusted daily according to the quotation at the London Exchange Market, and their value in U.S. dollars is converted into taka using the market exchange rate. SDR holdings are revalued daily, while other IMF accounts are revalued monthly using the IMF records and the taka/SDR cross rate.

BB’s credit to government takes the form of special treasury bills, whose interest is paid every 90 days. These bills are booked at their nominal value. The interest accrued in the meantime is not recorded, because the ABD does not calculate it on a monthly basis. However, the ABD is currently moving from a cash basis to an accrual system. Only at the end of the financial year (June), the ABD calculates the interest accrued on the BB’s holdings of treasury bills. In contrast, since June 2004, banks are reporting their holding of treasury bills at cost plus accrued interest, that is, at fair value. Treasury bonds, which are long-term securities, are reported by banks7 at face value, either underestimating (in case of bonds sold at par that pay a coupon) or overestimating (if the bonds were sold at a discount) their claims on government.

2.4.2 Recording is done on an accrual basis

Banks records are on an accrual basis. Prior to June 2004, accrued interest on deposits was reported in other provisions not included elsewhere, and accrued interest on loans was reported together with other assets. The new return provides a special item for accrued interest on loans, sectorized into government, other public, and private. Accrued interest on deposits is reported in one annex table, with one total figure by economic sector for all types of deposits. Interest accrued on nonperforming loans is included as part of total accrued interest and also as a liability item (interest suspense account), which is basically correct.

Until February 2005, the BB had a cash-basis system that was adjusted manually at the end of the financial year to comply with international financial reporting standards. At the time of this ROSC mission (early March 2005), the ABD had not yet produced on an accrual basis the February 2005 balance sheet; hence, the latest monetary data disseminated by the GSD were still compiled on a cash basis.

2.4.3 Grossing/netting procedures are broadly consistent with internationally accepted standards, guidelines, or good practices

Assets and liabilities of the banks are presented gross, with no netting of claims of a particular transactor against the liabilities to that transactor or group of transactors. Accounts in overdraft are recorded and reported as advances to the corresponding sector. Loans are recorded gross, and provisions for loan losses are reported by banks as other liabilities.

In the BB’s balance sheet, accounts in overdraft are netted from the balance of other current accounts and not shown as a claim of the BB on the institution in overdraft.

When preparing the ODC survey and the monetary survey, the GSD cancels out all outstanding claims and liabilities between the depository corporations. Assets and liabilities in the monetary survey are aggregated by resident and nonresident positions.

3. Accuracy and reliability

3.1 Source data

3.1.1 Source data are obtained from comprehensive data collection programs that take into account country-specific conditions

Banks transmit their monthly returns to the SD on a hard copy. Data cover the headquarters and all branches of the institutions. With the current presentation, the collected data allow for in-depth analysis by instrument and economic sector, and for cross checks at individual and aggregated levels. The format of the returns would allow for an easy aggregation of the individual submissions into a spreadsheet, but because the submission is not electronic, the whole process becomes cumbersome. After receiving the hard copies, the GSD forwards them to the ISDD for data entering into the system and receives back the processed individual data—also in hard copies—for validation of the entries and consistency checks. When problems are detected, the GSD contacts either the banks to correct the mistakes or the ISDD to rectify wrong entries. After receiving the GSD approval, the ISDD generates the ODC survey and provides the GSD with a hard copy of the results.

The GSD directly produces the central bank survey, albeit also manually. The ABD provides hard copies of the weekly and monthly condensed balance sheets of the BB, as well as additional information on currency in circulation, correspondent accounts abroad, position of banks at the BB, BB’s loans to banks, other assets, other liabilities, etc. With this additional information, the GSD manually corrects and expands the condensed balance sheet and generates the sectoral balance sheet and the central bank survey.

The GSD reviews the adequacy of the return forms and the annex tables on a yearly basis to adapt them to changing circumstances. The Guidelines are revised as needed. The latest comprehensive revision took place in June 2004, to introduce the recommendations of the IMF’s monetary and financial statistics mission. The SD keeps an updated register of the institutions belonging to the depository corporations sector and distributes it to the reporting units.

3.1.2 Source data reasonably approximate the definitions, scope, classifications, valuation, and time of recording required

The monetary survey is based on actual accounting records of the BB and almost all ODCs. Therefore, the GSD employs no sampling or estimations. The structure of the returns plus the annex tables permits an adequate sectorization and identification of the accounts of residents and nonresidents and a proper sectorization of the resident accounts.

3.1.3 Source data are timely

The ABD prepares the BB’s balance sheet with a time lag of approximately 20 days. This delay in producing such fundamental information is due to the lack of automatization of the accounting system. Transactions are recorded manually at the headquarters and at the nine branches (two in Dhaka and seven outside the capital city). Some of the branches in the interior send their balance sheets by mail, adding a delay of two to three days to their submission. Including the operations performed by the GSD, the central bank’s sectoral balance sheet and survey are finalized four or five weeks after the reference date, which is still within the requirements of the GDDS (two months) but untimely for an adequate policymaking decision process.

Banks submit their returns to the SD within four weeks after the end of the reference month. The processing of the data at the GSD takes ten additional days, and the drafting of the analytical notes that accompany the tables another ten days, producing a time lag for the dissemination of the monetary survey of about eight weeks.

3.2 Assessment of source data

3.2.1 Source data—including censuses, sample surveys and administrative records—are routinely assessed, e.g., for coverage, sample error, response error, and nonsampling error; the results of the assessments are monitored and made available to guide statistical processes

Because no automated internal checks are in place to reject the data when errors are detected, the GSD must perform all consistency checks manually. First, it verifies that assets and liabilities match in each individual return. It then investigates significant monthly movements in the balances of the accounts of the individual returns, which may be due to misreporting by the banks or errors in the data entry by the ISDD. Another consistency check performed by the GSD relates to the interbank position of money at call. The total of deposits reported in Table A of the annex is also checked against the total reported in the return. The reported stock of treasury bills and treasury bonds is verified against the BB’s debt registry. When discrepancies are detected, the GSD contacts the institutions to investigate the source of the discrepancy and correct the mistakes.

Staff of the GSD is comfortable about the quality of the data reported by ODCs, although it does not regularly visit the institutions to verify on-site the accuracy of the information submitted.

3.3 Statistical techniques

3.3.1 Data compilation employs sound statistical techniques to deal with data sources

Because the source data are processed manually, errors can occur when entering them. To eliminate these potential errors, the GSD carefully controls the processed information received from the ISDD.

Data reported by banks are complete, and the GSD does not estimate missing information. If any bank fails to report on time, the GSD has foreseen usinge its latest reported data and revising the published data the following month but has seldom been forced to apply this procedure.

3.3.2 Other statistical procedures (e.g., data adjustments and transformations, and statistical analysis) employ sound statistical techniques

The GSD converts foreign-currency-denominated accounts into taka using the market exchange rate provided by the BB. Furthermore, Table B of the annexes to bank returns breaks down foreign assets and liabilities by currency, allowing staff to check the accuracy of the translation from foreign into domestic currency.

For the monetary survey, the GSD calculates currency in circulation adjusting the data of the BB’s balance sheet with data on government currency in circulation (bills of small denomination and coins). Quarterly monetary data are seasonally adjusted using a moving average ratio.

3.4 Assessment and validation of intermediate data and statistical outputs

3.4.1 Intermediate results are validated against other information where applicable

The GSD checks foreign currency reserves of the BB against balance of payments data. This validation is not performed for the foreign assets of the banks. The GSD does not validate monetary data against data on financial market operations.

3.4.2 Statistical discrepancies in intermediate data are assessed and investigated

When discrepancies arise, the GSD contacts the relevant institution to solve the issue. One example of such cases is the holding of treasury bills, which is systematically cross-checked against the BB’s debt registry figures.

3.4.3 Statistical discrepancies and other potential indicators of problems in statistical outputs are investigated

The GSD keeps a file where it documents the most common detected mistakes. The file is consulted every month when assessing the accuracy of the reported data. Through courses at its Training Academy, the BB instructs officers of the banks and reduces potential factors of errors in their reporting.

3.5 Revision studies

3.5.1 Studies and analyses of revisions are carried out routinely and used internally to inform statistical processes (see also 4.3.3)

The main source of revision of monetary data is the technical assistance provided by the IMF to the BB. The recommendations of the last monetary and financial statistics mission have been almost fully implemented, with the corresponding revision of the methodology. Data were revised from June 2002 onwards for loans and holdings of government securities, and the revisions in the time series were indicated in the corresponding tables. The explanatory notes of the SD’s publications informed the public of the change in methodology.

4. Serviceability

4.1 Periodicity and timeliness

4.1.1 Periodicity follows dissemination standards

As recommended by the GDDS, the broad money survey and other monetary data (e.g., monetary base, interest rates) are published on a monthly basis in Economic Trends, which is also posted on the BB’s website.

4.1.2 Timeliness follows dissemination standards

Monetary statistics are disseminated with a two-month lag through Economic Trends and the BB’s website, complying with the recommendations of the GDDS.

4.2 Consistency

4.2.1 Statistics are consistent within the dataset

Monetary statistics are internally consistent. This includes the position of the BB with individual banks, and the interbank position of the individual institutions against each other. Consistency of monetary statistics against the consolidated balance sheet of other financial corporations is not being assessed, because no such sectoral balance sheet is being compiled.

4.2.2 Statistics are consistent or reconcilable over a reasonable period of time

Time series published in Economic Trends are consistent over time. Breaks in series are not identified in the corresponding tables but are addressed in explanatory notes.

4.2.3 Statistics are consistent or reconcilable with those obtained through other data sources and/or statistical frameworks

Foreign currency reserves of the BB are consistent with balance of payments data. This validation is not performed for the foreign assets of the ODCs. The GSD does not validate monetary data against data on financial market operations.

Every day, the BB sends to the MOF data on the government position, which should be an instrument to reconcile the monetary statistics with the data of government finance statistics. This notwithstanding, large discrepancies have been found in the data for domestic bank financing of government between the monetary accounts and the fiscal data.

4.3 Revision policy and practice

4.3.1 Revisions follow a regular and transparent schedule

The last published monetary data in Economic Trends are of a final nature but subject to revision the following month if errors are detected. These revisions were rare in the recent past.

The SBS-1 and NBCDs monthly returns are systematically reviewed and revised (sometimes only minor corrections) every year in June.

4.3.2 Preliminary and/or revised data are clearly identified

The public is not made aware that the last published monetary data in Economic Trends may be revised, neither through a note in the corresponding table nor in the explanatory notes. Tables posted on the BB’s website clearly identify when data are preliminary or revised.

4.3.3 Studies and analyses of revisions are made public (see also 13.5.1)

Studies and analyses of revisions are not made public.

5. Accessibility

5.1 Data accessibility

5.1.1 Statistics are presented in a way that facilitates proper interpretation and meaningful comparisons (layout and clarity of text, tables, and charts)

The BB disseminates monetary statistics through its monthly Economic Trends and the Bangladesh Bank Quarterly, both posted on the BB’s website. Detailed information pertaining to commercial banks, such as deposits by type of accounts, sectoral distribution of loans, etc., is published in the quarterly Scheduled Bank Statistics, while the Bangladesh Bank Bulletin contains quarterly monetary data. Both publications can be received by subscription but are not available on the BB’s website.

The tables in Economic Trends, Bangladesh Bank Bulletin, Bangladesh Bank Quarterly, and Scheduled Banks Statistics present the information clearly, with a level of disaggregation that meets different users’ needs. Explanatory and methodological notes are included in the first part of the publications, and analysis of macroeconomic trends or the respective tables accompany the monetary data. The Bangladesh Bank Bulletin presents quarterly data seasonally adjusted.

5.1.2 Dissemination media and format are adequate

Economic Trends and Bangladesh Bank Quarterly are posted on the BB’s website even before the printed copies are ready for distribution. Users can access longer time series upon request by contacting the GSD. Bangladesh Bank Bulletin and Scheduled Banks Statistics can be obtained only as hard copy by subscription.

5.1.3 Statistics are released on a preannounced schedule

Although the SD has an internal production cycle with specified deadlines for each stage of the process, which are normally met, there is no preannounced release calendar. The publications are not released on the same date every month, and delays have happened owing to unforeseen circumstances or holidays.

5.1.4 Statistics are made available to all users at the same time

Prior to their release to the general public, monetary statistics are made available to the MPD, for information purposes to the Minister of Finance, and for surveillance purposes to the IMF Resident Representative Office. The press is not briefed in advance when monetary statistics are disseminated.

5.1.5 Statistics not routinely disseminated are made available upon request

Upon request, the GSD provides nonpublished (but nonconfidential) information at a greater level of disaggregation.

5.2 Metadata accessibility

5.2.1 Documentation on concepts, scope, classifications, basis of recording, data sources, and statistical techniques is available, and differences from internationally accepted standards, guidelines, or good practices are annotated

Economic Trends, Bangladesh Bulletin, and Scheduled Banks Statistics contain an initial section with methodological notes on the tables and the terms and conditions under which monetary statistics are compiled. Metadata on monetary statistics are not available on the BB’s website but only through the IMF’s DSBB.

5.2.2 Levels of detail are adapted to the needs of the intended audience

The level of disaggregation contained in the BB’s publications is sufficient to meet different audiences’ needs.

5.3 Assistance to users

5.3.1 Contact points for each subject field are publicized

All publications list their editorial committees and the unit responsible for their edition. They identify the person responsible for the publication, indicating the position and e-mail address. The SD has not produced yet any material to make the general public aware of the significance of monetary statistics.

5.3.2 Catalogs of publications, documents, and other services, including information on any charges, are widely available

A catalog of the BB’s publications with their subscription prices is available on the Bank’s website.

Table 3.

Bangladesh: Data Quality Assessment Framework (July 2003): Summary of Results for Monetary Statistics

(Compiling Agency: Bangladesh Bank)

article image
article image

IV. Balance of Payments Statistics

0. Prerequisites of quality

0.1 Legal and institutional environment

0.1.1 The responsibility for collecting, processing, and disseminating the statistics is clearly specified

The Bangladesh Bank (BB) is responsible for the collection and dissemination of balance of payments statistics. The responsibility is established by tradition because the existing legal framework does not clearly designate the compilation of balance of payments statistics as the responsibility of the BB. The compilation and publication of the balance of payments data is based on the Bangladesh Bank Order, 1972, amended as of March 10, 2003, which requires banks to provide the BB with information as may be deemed necessary for carrying out the purpose of this Order. Under Chapter VI, Article 69, BB has the power to request from any natural person or juridical person, as well as the government, any data or information necessary for the proper execution of its duties. Confidentiality of reported data is assured under the Order by Chapter VI, Article 78. With regard to dissemination of data, the Order (Chapter III, Article 39) states that the Bank “shall compile and publish in such a manner and at such times as the Government may direct, a consolidated statement from such information as may be received under this order.” As regards working arrangements, the Balance of Payments Division (BOPD), in collaboration with the Foreign Investment and External Debt Division (FIED) in the Statistics Department of the BB, carries out the responsibility for balance of payments compilation and dissemination.

0.1.2 Data sharing and coordination among data-producing agencies are adequate

The Statistics Department receives data from various departments within the BB and institutions outside the BB. A formal institutional structure combines agencies working in the fields of foreign direct investment (FDI), balance of payments, and other statistics.

The Balance of Payments Projections Committee comprises the BB, the Bangladesh Export Processing Zones Authority (BEPZA), the Bangladesh Garments and Manufacturers Association of Exporters, the Export Promotion Bureau (EPB), the External Relations Division (ERD) of the Ministry of Finance (MOF), the Ministry of Commerce, and the Planning Commission. It generally meets twice a year to deliberate upon balance of payments issues and to make projections over the next one to two years. It is chaired by the Economic Advisor of the BB and assisted by senior officials of the BOPD. The Bangladesh Bureau of Statistics (BBS) is not involved in surveys conducted by the BB, which has organized its own surveys. The BB designs the report form, organizes the survey, and processes the results by itself.

Procedures are in place for the effective and timely flow of source data, particularly between the BB and the authorized dealers in foreign exchange (i.e., commercial banks) in support of the banking transactions records collected through the foreign exchange transactions reporting system. Balance of payments data are routinely made available to users and data-producing agencies. The exchange of data between official agencies and the BB is free of charge. In general, the BB has no need for data produced by the BBS for balance of payments purposes. Occasionally, consultations between the BBS and the BB occur when issues warrant. Monthly balance of payments are provided to the MOF and to the Ministry of Planning in a standard format.

0.1.3 Individual reporters’ data are to be kept confidential and used for statistical purposes only

The confidentiality of individual respondent data collected by the BB is guaranteed under the Bangladesh Bank Order, 1972, amended as of March 10, 2003. Article 78 specifies that every employee shall preserve and aid in preserving secrecy with regard to all matters relating to the affairs of the BB not published by it, and with regard to all matters relating to financial or monetary affairs of any institution, person, body of persons, any government, or authority, whether in Bangladesh or outside, that may come to his knowledge in the performance of his duties. Employees, if found guilty, face punishment of fines or imprisonment for a term which may extend to six months. The report form used in a major survey of foreign direct investment enterprises conducted by the BB explicitly informs the respondents that data obtained would be used for statistical purposes and kept confidential.

Access to individual data is restricted only to the relevant staff. Unauthorized access to individual data is prevented by using passwords to access computer files. In addition, confidentiality of data is properly guarded during storage and during the process of destruction, while the premises are appropriately secured. No special aggregation rules are employed to prevent residual disclosure, but there is a general review by staff of all prepared data aimed at preventing the disclosure of individual data. Unit records are generally not made available except to specific users (with concurrence of management), and when these unit records are made available, confidentiality requirements are imposed.

0.1.4 Statistical reporting is ensured through legal mandate and/or measures to encourage response

Under the Bangladesh Bank Order, 1972, amended as of March 10, 2003, Article 69, the Research Department and the Statistics Department of the BB shall have the power, subject to the rights and provisions guaranteed under other laws, to request from any natural or juridical person, as well as the Government, any data or information necessary for the proper execution of its duties. The commercial banks are also required to provide to the BB such information as may be deemed necessary for carrying out the purposes and object of this Order as the BB may direct. The Guidelines for Foreign Exchange Transactions, 1996, Volumes I and II, and circulars issued by the BB govern the administration of the foreign exchange control in Bangladesh. Authorized dealers provide monthly information to the BB, guided by the Code List for Reporting of External Transactions by the Authorized Dealers, 2000, issued by the Foreign Exchange Policy Department of the BB. The legal mandate for the BB to collect FDI and related statistics is based on Article 69 of the Bangladesh Bank Order, 1972. The foreign direct investment survey report form is accompanied by general notes, explaining the purpose of the survey, the legal basis, the confidentiality of the data provided, the period in which the report form needs to be submitted, and very detailed notes on definitions and terminology used in the survey report form. For assistance, the respondent can contact the BB by telephone, telefax, or e-mail.

At the level of the banks and their clients, the legal framework does not obligate clients to provide information, and the issue of penalties does not arise. At the level of the commercial banks and the BB, there is an obligation to respond and penalties are applicable. Article 36 of the Bangladesh Bank Order, 1972 imposes penalties for bank reporters who do not comply. Such penalties are only imposed in extreme cases as the BB resorts first to administrative procedures (verbal or written warnings). The collection activity is consistent with the legal authority. At the level of surveys, the BB can only request respondents to provide information, and there is no obligation on the part of the respondents to comply. Consequently, penalties are not imposed.

Other institutions, such as BEPZA, the Bangladesh Board of Investment, and the EPB collect statistics under the BEPZA Act 1980, the Investment Board Act, 1989, and the EPB Ordinance, 1977, respectively. The range of statistics is limited only to data on goods exports and imports and the registration of investment proposals.

In general, BB takes into account the response burden, but requesting information that cannot be derived easily from the accounting system of the respondent is difficult to avoid in practice. From time to time, the BB reviews the returns to further rationalize the data collection process. For surveys, the BB actively helps the respondents informally, through person-to-person contact, by telephone, etc. For the bank reporting system (International Transactions Reporting System or ITRS), the Bangladesh Bank Training Academy provides training to staff of commercial banks in classification definitions and other procedures of recording external transactions within the framework of the ITRS.

0.2 Resources

0.2.1 Staff, facilities, computing resources, and financing are commensurate with statistical programs

The Statistics Department has the BOPD and the FIED, which are in charge of compiling balance of payments statistics. A general manager heads the department, and each division is overseen by a deputy manager. The BOPD and FIED comprise nine and five sections respectively. Currently, 23 officials work in the two divisions, of which six are in the FIED. The educational qualification entry level in the Statistics Department is a master’s degree in statistics. Several members have also attended balance of payments methodology courses organized by the IMF. Overall, the number of staff is adequate to perform the required tasks. The staff members are qualified and experienced. A core staff with adequate training is maintained, and staff turnover is manageable. Salary levels follow the structure of government pay. Staff perceive that their salary levels are adequate, given the resources available to the BB, but these levels are not competitive with the private sector.

Computing resources are sufficient for existing balance of payments statistical programs, although not every staff member has his own computer (the BOPD has four computers). However, timeliness of completing the tasks could be improved if more computer resources were made available to the BOPD. The Information Systems Development Department (ISDD) aggregates ITRS data reported by approximately 700 commercial banks. The hardware to facilitate the efficient collection and processing of data and the management of databases is adequate, and the software used for compiling and analyzing statistical series is effective. Adequate protection is provided for the computer resources through the provision of emergency back-up systems and updating facilities. The BB plans to introduce electronic reporting in July 2005.

Office buildings provide adequate working facilities, and office furniture and equipment are in most cases adequate to perform the required tasks. Transportation arrangements for data collection are in place and utilized when required.

Funding for compiling statistics is reasonably secured for the identified needs of the statistical program. Technical assistance is sometimes required from international agencies and donors. Budgets are prepared at the central level. At the moment, forward planning for statistical development is not done.

0.2.2 Measures to ensure efficient use of resources are implemented

Staff performance reviews are conducted annually, but the promotion system is largely based on seniority as measured by the length of service. A system of fast-track promotions is also in place. Even though no formal process is in place to seek efficiencies through periodic reviews of work processes, management is aware of the need for efficiency. Instead, efficiencies are sought through informal discussions. There have been one or two instances in the past when outside expert assistance was sought to evaluate statistical methodologies and compilation systems. For instance, about three years ago assistance was sought to investigate the causes of the large errors and omissions item.

0.3 Relevance

0.3.1 The relevance and practical utility of existing statistics in meeting users’ needs are monitored

No structured and periodic process of consultation reviews the usefulness of existing statistics or identifies emerging data requirements. However, data users are consulted and/or kept informed on specific aspects through newsletters, publications, and website publications (e-mail address provided). In the BB publications, users are asked for their comments, suggestions, or queries. On rare occasions, the BB does receive feedback from users. No official of the Statistics Department has participated in any statistical meeting and seminar organized by professional, regional, or international organizations in the recent past.

0.4 Other quality management

0.4.1 Processes are in place to focus on quality

Management is sensitive to all dimensions of data quality. Local staff training programs are organized to emphasize the importance of quality and to give staff an understanding as to how quality can be achieved or improved. As noted earlier, the BB Training Academy also provides training to staff of commercial banks, aimed at improving the quality of the ITRS data. In addition, an instruction booklet with a detailed explanation of the standard components of the balance of payments has strengthened the ITRS. After returning from training courses, staff are asked to provide a presentation of what they have learned. A World Bank-funded Central Bank Strengthening Project is currently under way to improve quality through the development of a data warehouse.

0.4.2 Processes are in place to monitor the quality of the statistical program

Informal monitoring processes are in place to inform managers on the quality achieved for ongoing statistical activities. Compiling areas have access to expert guidance on the quality of their statistics and on strategies for improving data production. Periodic reviews are also undertaken to identify steps necessary to maintain quality requirements. Fortnightly meetings are chaired by the Economic Advisor and attended by the General Managers/Deputy General Managers in each department. In these meetings, quality issues and feedback from users are discussed.

0.4.3 Processes are in place to deal with quality considerations in planning the statistical program

No formal work program planning process addresses quality improvements identified during ongoing monitoring and periodic reviews. However, management is becoming gradually aware of the importance of the trade-offs among the dimensions of quality (e.g., resources availability, timeliness, and accuracy/reliability). Implicitly, staff has a keen awareness that in order to improve the timeliness (or when data are urgently needed by management) estimates need to be made on the best judgment possible. Feedback from users on quality standards and on new and emerging data requirements is rare.

1. Assurances of integrity

1.1 Professionalism

1.1.1 Statistics are produced on an impartial basis

Although no law or formal provision supports professional independence of the BB, traditions and a culture of professionalism date back well into the past. The BB does not have full operational autonomy from the MOF, and the legal environment does not protect the BB’s board directors against arbitrary removal. The Governor is appointed for a period of four years and is eligible for reappointment. Hiring policies seek out academically qualified statisticians with the appropriate expertise. Formal training in methodology and compilation methods is acquired through courses conducted by regional and international organizations, especially the IMF. No on-the-job training is in place, but professional literature is available and accessible to interested staff. Research and analysis are encouraged, and staff get credits for publications and participation in various discussion fora.

1.1.2 Choices of sources and statistical techniques as well as decisions about dissemination are informed solely by statistical considerations

The choice of source data (e.g, data collection through surveys, banking records, and administrative records, etc.) is based solely on statistical considerations. Similarly, decisions to disseminate data are based solely on statistical needs and economic considerations as well as the need of users.

1.1.3 The appropriate statistical entity is entitled to comment on erroneous interpretation and misuse of statistics

The BB provides explanatory material in its publications and briefings to the media and comments when its statistics are misinterpreted or misused. Instances of misinterpretation of statistics are generally dealt with through press releases. The BB management monitors the media coverage through the Executive Management Team (EMT), and the relevant department addresses any instances of misinterpretation. The EMT is headed by the Governor, and its members are the Deputy Governors and Executive Directors. Its aim is to deal with all aspects of management.

1.2 Transparency

1.2.1 The terms and conditions under which statistics are collected, processed, and disseminated are available to the public

In general, information is not available to the public about the terms and conditions under which statistics are collected, processed, and disseminated, except the report form used in the foreign direct investment survey. As noted before, in the foreign direct investment survey, the respondent is informed of the terms and conditions under which the survey is conducted. The survey report form for embassies and the foreign exchange control forms do not contain such information. The publications Economic Trends and Balance of Payments also do not contain information on the terms and conditions under which statistics are collected, processed, and disseminated to the public.

1.2.2 Internal governmental access to statistics prior to their release is publicly identified

Data are provided to the MOF and, occasionally, to other senior officials before they are released to the public. However, the public is not aware of internal government access to statistics prior to their release to the public. This practice is disclosed in the GDDS metadata of Bangladesh posted on the IMF’s Data Standard Bulletin Board.

1.2.3 Products of statistical agencies/units are clearly identified as such

Balance of payments data published in the BB’s publications are clearly identified by name, logo, and insignia. In the case of joint publications, the part attributable to other agencies is identified. The BB does not require attribution when its statistics are used for quotation or reproduction.

1.2.4 Advanced notice is given of major changes in methodology, source data, and statistical techniques

Advance notice of major changes in methodology, source data, and statistical techniques is currently not provided, but recently explanations are provided as and when these changes have occurred. The BB is reviewing procedures for providing advance notice of release.

1.3 Ethical standards

1.3.1 Guidelines for staff behavior are in place and are well known to the staff

The Bangladesh Bank Staff Regulations, 2003 spells out, among other things, ethical standards in articles covering aspects such as conflict of interest, acceptance of gifts and foreign awards, conduct towards female colleagues, and declaration of movable and immovable property. Liquid assets need to be disclosed when required, and the declaration on property needs to be submitted every five years. New staff are made aware of ethical standards when they join the BB during orientation courses and desk training. Staff are also reminded periodically of ethical standards. Also, in accordance with Article 71 of the Bangladesh Bank Order, 1972, every officer of the BB shall be deemed to be a public servant within the meaning of section 21 of the Penal Code.

2. Methodological soundness

2.1 Concepts and definitions

2.1.1 The overall structure in terms of concepts and definitions follows internationally accepted standards, guidelines, or good practices

Balance of payments accounts are compiled in accordance with the principles and methodology recommended in the fifth edition of the IMF’s Balance of Payments Manual (BPM5). Consequently, the current account balance is in principle equal (with sign reversed) to the net capital and financial account balance. The double-entry system is applied as a basic principle, the financial account provides for separate recording of transactions in assets and liabilities, and the directional principle is employed in recording foreign direct investment flows. Deviations from concepts and definitions are kept under review. In 1999, the BB converted its balance of payments framework to that recommended by the BPM5.

2.2 Scope

2.2.1 The scope is broadly consistent with internationally accepted standards, guidelines, or good practices

The balance of payments covers in principle all current, capital, and financial transactions between residents and nonresidents. In practice, the coverage may be incomplete if data sources to measure these transactions are not available. The coverage is ensured for both the financial and nonfinancial sector. Nontrade activities of enterprises in export processing zones and transactions between residents and nonresidents settled through offshore banks are included. However, the financial transactions of offshore banks with nonresidents are not included. Data on nonmonetary gold imports are not available, and no estimates are made for smuggling and other illegal flows. For goods for processing, no distinction is made between goods being returned to the country of origin after processing or processed goods exported to third countries because data to permit such a distinction are not available. Deviations from the scope broadly consistent with international practices are kept under review.

2.3 Classification/sectorization

2.3.1 Classification/sectorization systems used are broadly consistent with internationally accepted standards, guidelines, or good practices

The classification/sectorization is in accordance with BPM5 guidelines, comprising (i) the current account, including the goods, services, income, and current transfers account, (ii) the capital account, and (iii) the financial account, including direct investment, portfolio investment, the reserve assets, and net errors and omissions. Institutional units are classified and attributed to the relevant sector. For instance, government guaranteed loans are attributed to the borrower, and the foreign assets and liabilities of public enterprises are attributed to other sectors. The four sectors are also clearly identified in line with BPM5, namely, the monetary authorities, comprising the BB; the general government sector; deposit money banks, comprising all commercial banks; and other sectors, comprising nonbank financial institutions, enterprises, and individuals. Deviations, if any, from the above classification systems are kept under review. The following describes the main accounts:

Goods: Goods exports are recorded on an f o.b. basis, and goods imports are recorded on a c.i.f. basis, but adjusted to an f o.b. basis by deducting freight and insurance charges on goods imports rendered by residents as well as nonresidents. The goods account also includes goods for processing, repairs on goods, and goods procured in ports by carriers.

Services: This account includes transportation, travel, communications, construction, insurance, financial services, computer and information services, royalties and license fees, and other business services. In the transportation account, a distinction is made by mode of transportation. The travel account provides details on personal travel and travel for business purposes.

Income: This account includes data on compensation of employees, interest on loans, reinvested earnings, and undistributed branch profits.

Transfers: This account includes private donations, technical assistance, workers’ remittances, grants in food and commodities, official project grants, and Japan Debt Relief grants.

Direct investment: This account includes long-term construction projects. Borrowing and lending, including debt securities and supplier’s credits between direct investors and direct investment enterprises, are included in other capital of foreign direct investment.

Other investment: This account includes trade credits, loans including the use of Fund credit, currency and deposits, and other assets and liabilities.

Reserves: This account includes monetary gold, SDRs, the reserve position in the Fund, and foreign exchange reserve assets.

2.4 Basis for recording

2.4.1 Market prices are used to value flows and stocks

Transactions included in the ITRS are settled at market prices. Goods exports and imports are valued f o.b. Financial instruments are valued on the basis of prices prevailing in the markets where they are traded. Grants in kind and foreign direct investment equity contributions in kind are recorded at values reported by donors and foreign direct investors respectively. Nontradable financial instruments are valued at nominal value. Valuation and other changes are excluded when transaction estimates are derived from stock data.

Transactions are converted to taka and U.S. dollars at the average exchange rate for the relevant period based on the number of working days in the period, using midpoint exchange rates. The value of the reserve assets positions are updated on a monthly basis. Gold is valued on the basis of the US$ value of the gold troy ounce and on the US$ exchange rate versus the taka. The value of the troy ounce is revalued once a year using the London market rate, while the US$ exchange rate versus the taka is revalued irregularly.

2.4.2 Recording is done on an accrual basis

The change-of-ownership principle governs the recognition of transactions and their time of recording. Merchandise exports are recorded on an accrual basis. Grants and foreign direct investment equity contributions are recorded at values reported by donors and foreign direct investor instead of market values. Service transactions are recorded on a cash basis.

2.4.3 Grossing/netting procedures are broadly consistent with internationally accepted standards, guidelines, or good practices

In accordance with international guidelines, the current account items are recorded gross, while items in the financial account are recorded net, separately for assets and liabilities. In the transfers account, entries are shown as offsets to transactions to which there is no quid pro quo.

3. Accuracy and reliability

3.1 Source data

3.1.1 Source data are obtained from comprehensive data collection programs that take into account country-specific conditions

The balance of payments data are compiled from a number of sources, including merchandise exports based on customs declarations from the EPB, banking transactions records, surveys of foreign direct investment enterprises and foreign embassies, and administrative records. Data sources for estimating goods for processing in conformity with BPM5 are inadequate, and adjustments to goods imports derived from certain returns are not available. Goods for processing represent a significant component of the balance of payments in Bangladesh, but processed goods returned to the country of origin could not be distinguished from processed goods exported to third country.

Under the banking transactions reporting system, commercial banks submit foreign exchange returns, and the BB compiles balance of payments data derived from these returns on a monthly basis, which are both comprehensive and timely. A guide containing instructions and a detailed description of balance of payments components is provided to all commercial banks, facilitating the appropriate classification of external transactions. The bank reporting system has no threshold in terms of classification, but reporting instructions request explanations about transactions exceeding a certain amount to enhance the proper classification of large transactions. In addition, the reporting instructions do not permit netting and emphasize the need to minimize the number and the value of transactions that cannot be classified. Finally, the bank report forms are designed to make completion of the form easy and appropriate for computer processing.

Twice a year, the BB surveys foreign direct investment enterprises. Survey forms are reviewed periodically to take account of changing circumstances. Thus far, only minor changes have been made and no pretests were necessary to test their effectiveness.

Data from administrative records are adequate, and their gaps in terms of detail and coverage are known. Efforts are made to account for those gaps. The available data sources are broadly sufficient to compile balance of payments statistics. The data collection programs are sufficiently open and flexible to provide for new developments in sources. To further this end, periodic meetings, if required, are being held with financial market participants and the business community to identify new developments. International standards, guidelines, and practices are monitored for changes that need to be incorporated in the balance of payments compilation system.

Goods: Data on merchandise exports f.o.b. are based on customs declarations. Data on merchandise imports, recorded on a c.i.f. basis, are primarily derived from the exchange record and supplemented with (i) data on imports funded by external official debts and grants and (ii) data on goods imports provided by BEPZA. The latter covers goods imported by enterprises operating in the export processing zones. To adjust imports on a c.i.f. basis to an f.o.b. basis, it is assumed that insurance and freight represent 10 percent of the f.o.b. value.

Services: With the exception of freight and insurance charges on merchandise imports, data on all other services are obtained from exchange control records.

Income: Data on compensation of employees are derived from surveys of foreign embassies, and those on investment income are derived from foreign exchange control records. Data on income are supplemented by data on interest payments on loans provided by the ERD of the MOF and data on interest received by the BB.

Transfers: Data on current and capital transfers are derived from exchange control records and from source data compiled by the ERD. The latter compiles the data on the basis of internal accounting procedures as well as reports from donors.

Financial account: Data on direct investment in Bangladesh and abroad are derived from enterprise surveys. Monetary and financial statistics on the external positions of resident units are used to derive estimates for the financial transactions of these units. Data on loans are derived from ERD reports. All other entries are derived from exchange control records. Data on international reserve assets are derived from the internal records of the BB.

3.1.2 Source data reasonably approximate the definitions, scope, classifications, valuation, and time of recording required

Source data from surveys and the ITRS are consistent with the definitions, scope, classifications, and time of recording of the balance of payments statistics employed by the BB. Administrative records provide reasonable approximations of the methodological requirements of the balance of payments. The BB is aware of timing and coverage differences, but adjustments to the data are difficult to make as the respondents cannot be reached. Efforts to partially adjust the data focus on making comparisons with similar data. Information on external debt is broadly in line with the recommendations of BPM5.

3.1.3 Source data are timely

The source data are timely because the data collection programs provide fixed deadlines. Respondents are made aware of these deadlines, and the BB employs rigorous follow-up procedures to ensure the timely receipt of respondents’ data.

3.2 Assessment of source data

3.2.1 Source data—including censuses, sample surveys and administrative records—are routinely assessed, e.g., for coverage, sample error, response error, and nonsampling error; the results of the assessments are monitored and made available to guide statistical processes

Verification of data pertains to the checking mechanism internal to the survey report forms used in the direct investment survey. The forms establish equalities, such that the closing position of the previous period should be equal to the opening position of the current period or that the opening position plus flows and other changes should be equal to the closing position. Queries are followed up by telephone and/or visits until they are resolved satisfactorily. As noted earlier, only minor changes have been made thus far.

In the ITRS, the BOPD edits, reviews, and clears statements submitted by banks before sending them to the ISDD. Any discrepancies are resolved with the reporting banks. As noted earlier, values exceeding a certain amount require a more detailed explanation and, if necessary, a confirmation with the banks. Sample surveys are not conducted to classify low-value transactions across the balance of payments components.

3.3 Statistical techniques

3.3.1 Data compilation employs sound statistical techniques to deal with data sources

Compilation procedures minimize processing errors such as coding, editing, and tabulation errors. Unusual values, if they can be identified, are not replaced or modified unless clearly required. Procedures for imputation and adjustment for nonresponse are soundly based. Sound estimation techniques are used to adjust data for missing observations. Discrepancies between the sum of monthly to quarterly data and the quarterly data to annual data are eliminated.

3.3.2 Other statistical procedures (e.g., data adjustments and transformations, and statistical analysis) employ sound statistical techniques

To adjust imports on a c.i.f. basis to an f.o.b. basis, it is assumed that insurance and freight represent 10 percent of the f.o.b. value. This rule of thumb has been not been reviewed for a number of years. Another adjustment is used to estimate technical assistance, currently estimated at 7 percent of project grants.

3.4 Assessment and validation of intermediate data and statistical outputs

3.4.1 Intermediate results are validated against other information where applicable

News media are monitored, and the information derived is checked with banking records. At present, export and import duties are not cross-checked with the exports and import totals, owing to nonavailability of information in bank returns.

3.4.2 Statistical discrepancies in intermediate data are assessed and investigated

The BB regularly assesses reported data on investment income receipts and payments in relation to the stock data in the international investment position. The reported financial flows are also reconciled with changes in the corresponding stock data collected for external debt and other elements in the international investment position. In some cases, the BB also reviews discrepancies between the merchandise trade data and the associated financial flows from the ITRS.

3.4.3 Statistical discrepancies and other potential indicators of problems in statistical outputs are investigated

The BB monitors developments in the net errors and omissions item through cross-checks between entries in the current, capital, and financial accounts. It also reconciles bilateral data for selected trade and other current and financial account items with the principal trading partners; large differences are investigated. Differences in concepts and compilation methods are identified, but the revision of data or reconciliation tables cannot be worked out, owing to the lack of detailed information on the compilation procedures in other countries.

3.5 Revision studies

3.5.1 Studies and analyses of revisions are carried out routinely and used internally to inform statistical processes (see also 4.3.3)

Over time, initial estimates are assessed against revised or final estimates. These assessments include studies of the frequency of revisions and its direction and magnitude as and when necessary. The BB uses findings from revision studies to define the optimal revision cycle or to refine the preliminary data and data collection programs for subsequent periods.

4. Serviceability

4.1 Periodicity and timeliness

4.1.1 Periodicity follows dissemination standards

The BB compiles balance of payments data on a monthly, quarterly, and annual basis, thus exceeding the periodicity requirement of the General Data Dissemination System (GDDS) and meeting the requirement of the Special Data Dissemination Standard (SDDS).

4.1.2 Timeliness follows dissemination standards

The timeliness of balance of payments data complies with the GDDS recommendation and has reached the level of the SDDS. The BB compiles balance of payments data within six weeks after the end of the reference period and disseminated within two months after the end of the reference period.

4.2 Consistency

4.2.1 Statistics are consistent within the dataset

Balance of payments data are internally consistent, and annual data are the sum of the quarterly data. In a similar vein, quarterly data are the sum of monthly data. Concepts, definitions, and classifications for producing the monthly, quarterly, and annual data are the same.

4.2.2 Statistics are consistent or reconcilable over a reasonable period of time

Balance of payments data are consistent over time with time series available over a period of at least five years. When the BB introduces changes in source data, methodology, or techniques, it reconstructs historical time series as far back as reasonably possible. Detailed methodological notes accompany the balance of payments tables addressing some but not all of these changes. Unusual changes in economic trends, though not the underlying reasons, are highlighted in the commentary included in the balance of payments publication,.

4.2.3 Statistics are consistent or reconcilable with those obtained through other data sources and/or statistical frameworks

Balance of payments data are reconcilable or consistent with data from various data sources. Data on goods exports are reconciled with the customs data. Data on goods imports are difficult to reconcile with the customs data because data sources to adjust these data to a balance of payments basis are not available. The balance of payments data are also largely consistent with the national accounts and the monetary and financial statistics. Finally, balance of payments components comprising external debt data are largely consistent with the corresponding debt stocks.

4.3 Revision policy and practice

4.3.1 Revisions follow a regular and transparent schedule

Not formally established, data revision cycle does not necessarily follow a predictable pattern. Data are revised when corrections have been implemented or when more complete returns are received from other depository corporations. Data revisions are indicated by footnotes. The revision cycle is not made known to the public. Revisions occurring outside the regular revision cycle, for instance when new data sources have been developed or when errors have occurred, are made known to the public.

4.3.2 Preliminary and/or revised data are clearly identified

Preliminary data are identified in the Annual Report but not in the Balance of Payments, since the data published here are final. Preliminary and/or final data are not identified in Economic Trends. Users are informed when data are revised at the time of dissemination.

4.3.3 Studies and analyses of revisions are made public (see also 3.5.1)

Users are informed of results and studies of the revision to the statistics.

5. Accessibility

5.1 Data accessibility

5.1.1 Statistics are presented in a way that facilitates proper interpretation and meaningful comparisons (layout and clarity of text, tables, and charts)

Balance of payments data are presented to facilitate proper interpretation and meaningful comparisons with previous periods. Monthly data are presented summarized. Summarized quarterly data are presented in accordance with BPM5 format, as are detailed annual data. Data are compiled and published in millions of taka. In the Annual Balance of Payments, commentaries on current developments and definitions facilitate the understanding of the balance of payments. There are no supporting tables in the annual publication. Detailed regional balance of payments data are also disseminated.

5.1.2 Dissemination media and format are adequate

Balance of payments data are published in the Bangladesh Bank Quarterly, Bangladesh Bank Bulletin, Economic Trends, the Annual Report, and the Annual Balance of Payments.

Balance of payments data are also disseminated on the BB’s website (http://www.bangladesh-bank.org).

5.1.3 Statistics are released on a preannounced schedule

Balance of payments data are not released according to an advance release calendar.

5.1.4 Statistics are made available to all users at the same time

Data are provided to the users simultaneously. The public is not aware of balance of payments statistics being released, but users are aware of when balance of payments data should become available. The BB sometimes receives enquiries about the next release of balance of payments data

5.1.5 Statistics not routinely disseminated are made available upon request

Unpublished data are made available upon request. Customized tabulations can be provided. The availability of additional statistics and the procedures to obtain additional statistics is not known. However, the Economic Trends does note to whom queries could be made.

5.2 Metadata accessibility

5.2.1 Documentation on concepts, scope, classifications, basis of recording, data sources, and statistical techniques is available, and differences from internationally accepted standards, guidelines, or good practices are annotated

Documentation of methodology, sources used in preparing statistics, and an explanation of the types of transactions included in each of the data categories are published in Balance of Payments. Methodological changes are reported to the public through notes in the monthly Economic Trends and the quarterly Bangladesh Bank Bulletin. The BB also provides methodological information for publication in the IMF’s Balance of Payments Statistics Yearbook and on the GDDS website. No information is provided on the metadata for surveys.

5.2.2 Levels of detail are adapted to the needs of the intended audience

Additional information on metadata is made available upon request. Background papers and working documents are not available to the users. General information, such as brochures, is also not available.

5.3 Assistance to users

5.3.1 Contact points for each subject field are publicized

An e-mail and a mailing address are provided in the Economic Trends. A website address is also provided in all the publications of the BB.

5.3.2 Catalogs of publications, documents, and other services, including information on any charges, are widely available

A list of publications is available and updated each year in all the publications of the BB, except the Bangladesh Bank Quarterly and the Annual Report. The prices of the publications are clearly disclosed, and mailing, fax, and e-mail addresses are provided for obtaining these publications.

Table 4.

Bangladesh: Data Quality Assessment Framework (July 2003): Summary of Results for Balance of Payments Statistics

(Compiling Agency: Bangladesh Bank)

article image
article image
article image

APPENDIX I Summary of the General Data Dissemination System (GDDS)

Data coverage, periodicity, and timeliness

Dissemination of reliable, comprehensive, and timely economic, financial, and socio-demographic data is essential to the transparency of macroeconomic performance and policy. The GDDS contains specific recommendations concerning coverage, periodicity, and timeliness for comprehensive frameworks as well as for data categories and indicators.

Quality

Data quality must have a high priority. Data users must be provided with information to assess quality and quality improvements. The GDDS recommends:

  • dissemination of documentation on methodology and sources used in preparing statistics; and

  • dissemination of component detail, reconciliations with related data, and statistical frameworks that support statistical cross-checks and provide assurance of reasonableness.

Integrity

To fulfill the purpose of providing the public with information, official statistics must have the confidence of their users. In turn, confidence in the statistics ultimately becomes a matter of confidence in the objectivity and professionalism of the agency producing the statistics. Transparency of practices and procedures is a key factor in creating this confidence. The GDDS, therefore, recommends:

  • dissemination of the terms and conditions under which official statistics are produced, including those relating to the confidentiality of individually identifiable information;

  • identification of internal government access to data before release;

  • identification of ministerial commentary on the occasion of statistical releases; and

  • provision of information about revision and advance notice of major changes in methodology.

Access to the public

Dissemination of official statistics is an essential feature of statistics as a public good. Ready and equal access by the public are principal requirements. The GDDS recommends:

  • dissemination of advance release calendars; and

  • simultaneous release to all interested parties.

Plans for improvement

The GDDS recommends that plans for improvement be developed for all areas in which shortcomings exist and that these plans be disseminated.

The GDDS also recommends that any needs for assistance be identified in the metadata. This may also be helpful for donors and technical assistance providers to prioritize their activities.

For each participating member country, the GDDS metadata provide descriptions of the dimensions listed above, together with plans for improvement and needs for assistance. This information is posted on the DSBB; participating countries are encouraged to also post the metadata on their national websites.

Source: Guide to the GDDS, March 2002: http://dsbb.imf.org

APPENDIX II

Data Quality Assessment Framework—Generic Framework (July 2003 Framework)

article image
article image
article image

APPENDIX III Summary of Results of Survey of Data Users

The mission, with the assistance of the office of the IMF Resident Representative in Dhaka, conducted a survey of users of macroeconomic statistics in Bangladesh. The survey complemented the mission’s assessment of the quality of the statistics. Users were asked to evaluate and comment on the coverage, periodicity, and timeliness of the datasets in the ROSC assessment, as well as the dissemination practices of the compiling agencies. Questionnaires were sent to 39 users, and 12 responses were received from users representing international organizations, academia, research institutes, and banks.8 While in Dhaka, the mission held discussions with a selection of users.

On a scale of 1 to 5, with 5 being “excellent” and 1 being “poor,” the quality of the various data sets was rated as average or slightly below average. In most areas, the users considered the quality of Bangladesh statistics to be similar to that of other countries in the region.

The following key points emerged from both the survey and discussions held:

  • Coverage and level of detail in national accounts statistics could be improved.

  • Timeliness for the dissemination of statistics could be improved, particularly in the areas of national accounts and price statistics. Periodicity was generally considered satisfactory, except for the national accounts statistics.

  • ■ Not enough information appears about the calendar of publication and revision policies. Some users felt that the information about revisions is too brief and not clearly explained.

  • ■ Users can easily access the official statistics, with government agencies viewed as having better access to statistics. The level of detail of the metadata was seen as insufficient.

  • ■ The methodology of statistics was viewed as almost satisfactory, while the accuracy of almost all data sets is considered to need improvement.

  • ■ Several users indicated the need to improve transparency and build confidentiality in the official statistics.

Table 5:

Bangladesh—Results of Users’ Survey

article image
article image
article image
NA = National Accounts; Prices refers to: CPI (Consumer Price Index) and PPI (Producer Price Index); GFS = Government Finance Statistics; Monetary = Monetary Statistics; and BOP = Balance of Payments Statistics
1

Amended through the Bangladesh Bank (Amendment) Act, 2003.

2

Equivalent to 8,000 U.S. dollars as of March 2005.

3

Equivalent to 400 U.S. dollars as of March 2005.

4

The total number of positions assigned to the SD was 63, but 12 were vacant.

5

Commercial banks correctly report the accounts of the BR and the BTTB as part of the public nonfinancial corporations sector. The BPO is included within the central government.

6

Without agreement of the MOF and corresponding change in fiscal statistics, the reclassification of these public enterprises in the monetary statistics will create inconsistency between monetary and fiscal statistics and will affect the monitoring of quantitative targets under Bangladesh’s arrangement with the IMF.

7

The BB does not hold treasury bonds.

8

In spite of repeated follow-ups, the response rate was low. However, the IMF Resident Representative confirmed that those who responded are the avid users of statistics in Bangladesh and that there is generally a low level of statistical awareness in the country.

Bangladesh: Report on the Observance of Standards and Codes—Data Module
Author: International Monetary Fund