Pakistan: Report on the Observance of Standards and Codes (ROSC)—Data Module
Author: International Monetary Fund
Publication Date:
07 Dec 2004
eISBN:
9781451830613
Language:
English
Keywords:
ISCR; CR; information; government; procedure; SBP; WPI; data quality; quality assessment; data-producing institution; hard copy; data user; Government finance statistics; Monetary statistics; Data dissemination; contact information
Download PDF (744.8 KB)

This Report on the Observance of Standards and Codes (ROSC) provides a review of Pakistan’s data dissemination practices against the IMF’s General Data Dissemination System (GDDS), complemented by an in-depth assessment of the quality of the national accounts, consumer and wholesale price indices, and government finance, monetary, and balance-of-payments statistics. Pakistani authorities are strongly committed to adhering to internationally accepted standards and good practices, as demonstrated by their recent participation in the GDDS. Nevertheless, the mission identified shortcomings in some statistical practices and databases.

Abstract

This Report on the Observance of Standards and Codes (ROSC) provides a review of Pakistan’s data dissemination practices against the IMF’s General Data Dissemination System (GDDS), complemented by an in-depth assessment of the quality of the national accounts, consumer and wholesale price indices, and government finance, monetary, and balance-of-payments statistics. Pakistani authorities are strongly committed to adhering to internationally accepted standards and good practices, as demonstrated by their recent participation in the GDDS. Nevertheless, the mission identified shortcomings in some statistical practices and databases.

I. Introduction

1. The data module of this Report on the Observance of Standards and Codes (ROSC) provides a summary assessment of practices on the coverage, periodicity, and timeliness of the data categories against the IMF’s General Data Dissemination System (GDDS).5 It is complemented by an assessment of the quality of national accounts, consumer and wholesale price indices, and government finance, monetary, and balance of payments statistics, using the Data Quality Assessment Framework (DQAF) developed by the IMF’s Statistics Department (STA). This report is based on information gathered during a staff mission from December 1-16, 2003,6 publicly available information, as well as an informal users’ survey conducted ahead of the mission.

2. Section II provides an assessment of Pakistan’s data dissemination practices against the GDDS. At the request of the authorities, their current data dissemination practices were also reviewed against SDDS requirements, focusing on the coverage, periodicity, and timeliness prescriptions of the data dimension. Some significant issues that need attention prior to Pakistan’s subscription to the SDDS are identified. Section III presents a summary assessment of the quality of the principal macroeconomic datasets, following the dataset-specific assessment frameworks. Finally, Section IV sets out recommendations for further improvements in Pakistan’s macroeconomic statistics.

II. Data Dissemination Practices and the GDDS: Current Dissemination Practices

3. Pakistan’s data dissemination practices are assessed against the GDDS. Where shortcomings are found, the focus is on the plans for improvements posted on the DSBB. Metadata were first posted on the DSBB in November 2003.

4. Macroeconomic statistics in Pakistan are mainly produced by three institutions: (i) the Federal Bureau of Statistics (FBS) is responsible for the national accounts, prices, other economic statistics; (ii) the State Bank of Pakistan (SBP) is responsible for monetary and balance of payments statistics; and (iii) the Ministry of Finance (MOF) is responsible for statistics on central and general government operations and central government debt.

5. Access to macroeconomic data and socio-demographic indicators is provided through several publications and at the following Internet websites:

Data dimension: coverage, periodicity, and timeliness

6. In general, Pakistan meets the GDDS recommendations for the comprehensive frameworks. However, there is room for improving the coverage of most datasets, particularly of national accounts. With respect to the periodicity and timeliness of core indicators, most GDDS recommendations are also met, with the exception of data on GDP and wages/earnings. The first GDP releases, which are published about 2 weeks before the end of the fiscal year, are based on 9 to 10 months’ observed data and two to three months’ projections. The first annual GDP data based on 12 month’s observations are disseminated with a lag of 11 months (vis-à-vis the recommended 6-9 months). This situation would be remedied with the publication of quarterly national accounts as presently in preparation. Annual data on wages/earning are not disseminated. Table 1 shows an overview of current practices regarding coverage, periodicity, and timeliness of data in Pakistan compared to the GDDS.

Table 1.

Pakistan: Overview of Current Practices Regarding Coverage, Periodicity, and Timeliness of Data Compared to the General Data Dissemination System

article image
article image
Italics indicate encouraged categories.

Quality dimension

7. The quality dimension of the GDDS is particularly important, given that the primary focus is on improvements in data quality over time. The assessment of quality is based on the dissemination of (i) documentation on methodology and data sources, and (ii) component detail and reconciliation with related data.

8. Summary documentation on the methodology for all macroeconomic statistics in Pakistan is available on the DSBB. However, only limited detailed documentation on concepts and methods is published by the data-producing institutions and the DSBB is not yet sufficiently publicized on their websites.

Integrity dimension

9. The GDDS recommends the disclosure of the legal framework for the collection, compilation, and dissemination of data, including the confidentiality of the collected data.

10. The terms and conditions under which most official statistics are compiled and disseminated in Pakistan provide a legal framework that supports, to a large extent, the integrity of the statistical system, but this information is not sufficiently publicized. Procedures on the government’s internal access to the data prior to public release are not always disseminated.

Access dimension

11. Dissemination of official statistics is an essential feature of statistics as a public good. Ready and equal access are principal requirements for the public, including market participants. The access dimension is based on two practices that facilitate ready and equal access to data: (i) dissemination of advance release calendars; and (ii) simultaneous release to all parties.

12. Advance release calendars are not in all cases provided to users, although all institutions follow internal fixed-term release schedules that are largely known by the public. Hardcopies of the data are released to selected users within the government as well as to the press shortly before they are published on the FBS website.

Plans for improvement

13. Plans for improving data quality are an integral part of the GDDS. The following comments are based on information available on the DSBB and comments provided by the authorities during the mission.

14. For the real sector statistics, focus is on expanding data sources and broadening the scope of the statistics. For prices, plans are to introduce COICOP and complete work on development of a PPI with base year 1999/2000. With respect to labor market statistics, the FBS is investigating the feasibility of compiling quarterly employment/unemployment data and of disseminating data on wages/salaries.

15. General plans for improvement of government finance statistics (GFS) include the Fiscal Responsibility Law, which has been already submitted to parliament for approval, and the Project for the Improvement of Financial Reporting and Auditing (PIFRA). The MOF also plans to develop and disseminate metadata and an advance release calendar for GFS. In the longer term, the MOF intends to adopt the Government Finance Statistics Manual 2001 (GFSM 2001).

16. In the area of monetary statistics, the SBP will continue to adopt the Monetary and Financial Statistics Manual (MFSM). In particular, the SBP plans to (i) disaggregate data on nonfinancial corporations and households within the private sector; (ii) collect data on nonbank financial institutions to expand the coverage of the monetary aggregates; (iii) and develop and disseminate metadata on the compilation of monetary statistics.

17. Regarding the external sector, the SBP plans to (i) continue the full implementation of the Balance of Payments Manual, fifth edition (BPM5); (ii) initiate plans to compile and publish data on the international investment position; and (iii) disseminate the metadata for the template on International Reserves and Foreign Currency Liquidity on the SBP website. The SBP also plans to develop advance release calendars for monetary and balance of payments statistics.

Special Data Dissemination Standard (SDDS)

18. Since Pakistan has expressed interest in subscribing to the SDDS in the near future, the current data dissemination practices were also reviewed against its requirements.7 The following points about the coverage, periodicity, and timeliness prescriptions of the data highlight significant issues to be addressed prior to Pakistan’s subscription to the SDDS:

  • In the real sector, all data categories meet SDDS coverage requirements, with the exception of the annual data on wages/earnings, which are not disseminated. With respect to periodicity and timeliness, the following data categories do not meet the requirements: (i) the first GDP releases, which are published about two weeks before the end of the fiscal year, are based on nine to ten months’ observed data and two to three months’ projections. The first annual GDP data based on 12 month’s observations are disseminated with a lag of 11 months (vis-à-vis the recommended 6-9 months). This situation would be remedied with the publication of quarterly national accounts as presently in preparation.; (ii) employment/unemployment data, which are compiled on an annual basis and disseminated with a lag of six months (vis-à-vis the recommendation for quarterly periodicity and timeliness); and (iii) data on wages/earnings (SDDS requires quarterly periodicity and timeliness).

  • In the fiscal sector, data on general government operations basically meet SDDS requirements on periodicity and timeliness. For coverage, even though the definition of general government excludes locally financed operations of district and local governments, their operations are immaterial. As regards data on the operations of the central government, they are only compiled on a quarterly basis and disseminated with a lag of two months (against the SDDS requirement of monthly periodicity and timeliness).

  • In the financial sector, data on the analytical accounts of the banking system, interest rates, and the share price index meet SDDS requirements. However, the analytical accounts of the SBP do not distinguish the main aggregates clearly. SBP data are disseminated only four weeks after the end of the reference month, against the required two weeks.8

  • In the external sector, most data categories (balance of payments, international reserves and foreign currency liquidity, merchandise trade, and exchange rate) meet the coverage, periodicity, and timeliness requirements of the SDDS. However, only partial and non-disseminated quarterly data are available on the international investment position (vis-à-vis the annual periodicity and semiannual timeliness required by the SDDS).

19. Appendix I in the accompanying document presenting the detailed assessments provides an overview of Pakistan’s current practices regarding coverage, periodicity, and timeliness of data compared to the SDDS.

III. Summary Assessment of Data Quality

20. Interest in assessing the quality of data derives from the need to complement the GDDS with more accurate and reliable data to support the surveillance of countries’ economic and financial policies. Against this background, the IMF’s Statistics Department (STA) has developed a tool based on international best practice to provide structure and a common language to assess data quality. The Data Quality Assessment Framework (DQAF) comprises a generic framework9 and a set of dataset-specific frameworks covering five dimensions of data quality—integrity, methodological soundness, accuracy and reliability, serviceability, and accessibility—and a set of prerequisites.10

21. An assessment of six macroeconomic datasets (national accounts, consumer price index, wholesale price index, and government finance, monetary, and the balance of payments statistics) was conducted, using the frame of reference provided by each dataset-specific DQAF. The application of this framework to the Pakistani statistical system is presented below, following the structure of the DQAF. Conclusions are presented in standardized summary tables that assess data practices in terms of their adherence to international best practice, using a four-part scale (Table 2 in this report and Tables 1-6 in the accompanying document).

Table 2.

Data Quality Assessment Framework: Summary Presentation of Results for Pakistan

article image
article image

22. Pakistan’s macroeconomic statistics and statistical base are considered broadly adequate by the IMF’s Middle East and Central Asia Department (MCD) for effective surveillance. Nevertheless, the mission identified shortcomings in some statistical practices that may detract from the accurate and timely analysis of economic and financial policies.

Prerequisites of quality

The “Prerequisites of Quality” in the DQAF identifies conditions within the agency in charge of producing statistics that have an impact on data quality. The elements within the category refer to the legal and institutional environment, resources, and quality awareness.

23. The legal and institutional environment for compiling official statistics is established under the General Statistics Act No. 69 of 1975. The Act does not state explicitly that the Federal Bureau of Statistics (FBS) is responsible for the national accounts or price statistics; however, this is implied by the official status of the FBS and the tradition of its being the only agency compiling such statistics in Pakistan. The coordination of statistical programs is a task of the National Statistical Council (NSC), in which major ministries, chambers of commerce, researchers, and other users of statistics are represented. The Act guarantees the confidentiality of all data and specifies penalties for the breach of confidentiality by staff. It also provides the FBS with the legal basis for collecting the requisite data from all national units; however, the agency has experienced difficulty in having enterprises respond to questionnaires. Even though the Act establishes fines for noncompliance, the amounts of the fines are not indexed and have become insignificant for many nonrespondents. Resources for compiling national accounts statistics are generally below the desired levels. Though the number of staff in the National Accounts Wing (NAW) is broadly adequate, the technical skills of the staff may not be conducive to achieving its development goals. Information technology resources are deficient; most available computers are obsolete. The staff resources dedicated to the compilation of price statistics are broadly adequate. While computer resources in the Karachi office are just sufficient for the current program, only a few field offices have computers to transmit the price information by electronic mail. Many offices transmit the information by fax or express mail delivery. The managers of the FBS are sensitive to all dimensions of data quality and recognize the tradeoffs within quality. Quality awareness is also evidenced by the convening of regular meetings among senior staff in the NAW to review work practices. Committees on national accounts and price statistics, comprising data providers and users, meet on a regular basis to review the statistical techniques and provide feedback on data quality issues.

24. The legal and institutional environment is largely supportive of government finance statistics (GFS). However, the responsibility for the compilation and dissemination of comprehensive government finance statistics by the Ministry of Finance is not well defined, which hampers the compilation of the fiscal aggregates with broader coverage than budgetary government. Arrangements for the supply of GFS revenue and expenditure aggregates by the Accountant General for Pakistan Revenue (AGPR) and financing aggregates by SBP are well established. Staff and computer resources in the MOF are adequate to produce GFS on a limited scale, but staff would benefit from further training in statistical methodologies. In AGPR, the main resource constraint is lack of computer automation of GFS processing, which significantly limits the range of GFS data that can be produced. Quality awareness should be strengthened. GFS compilers lack a user perspective and focus on processes rather than products. There are planned improvements to budget accounting standards and systems, which will also allow improvements to GFS, but there are no specific processes in place to deal with quality considerations.

25. The State Bank of Pakistan (SBP) has a legal and institutional environment that supports its responsibilities for compiling and disseminating monetary and balance of payments statistics. Even though the existing laws do not explicitly mandate the SBP to compile monetary and balance of payments statistics, the SBP has been compiling these data since its establishment in 1948. Data sharing arrangements within the SBP and with other data providers are satisfactory. The laws establish the confidentiality and nondisclosure of data, with penalties for noncompliance. Statistical reporting of the banks is ensured through the Banking Companies Ordinance Act of 1962 and official notices to encourage response. The financial and computing resources allocated to the existing statistical programs are adequate to perform the required tasks and are used efficiently. Human resources are, in general, adequate for work on monetary and balance of payments statistics; however, monetary statistics compilers require further training in statistical methodology. The SBP demonstrates quality awareness through its published “Vision Statement” and the Statistics Department’s business plan, which show that managers are sensitive to all dimensions of data quality. The Statistics Department’s Planning and Development Division addresses quality issues and provides guidelines on new and emerging data requirements.

Integrity

Integrity identifies features that support firm adherence to objectivity in the collection, compilation, and dissemination of statistics so as to maintain users’ confidence. Elements refer to the professionalism and ethical standards that should guide policies and practices, which should be reinforced by their transparency.

26. The terms and conditions for the recruitment and promotion of staff at the FBS follow the public service regulations of Pakistan and encourage professionalism. Staff are appointed on a permanent basis after meeting the necessary civil service entry requirements. The head of the FBS can only be appointed from the ranks of the staff and is eligible to remain in the post until retirement. Some users of official statistics expressed concern about the delays in appointing a Director-General for the FBS and the authorities are actively seeking to identify an appropriate candidate. Furthermore, the decision on the dissemination of the national accounts statistics rests with the FBS. The National Accounts Committee, comprising individuals from government agencies and private sector institutions, is responsible for reviewing and approving the accounts. This committee, which is chaired by the Secretary of the Statistics Division, focuses on providing updated source data, and may also recommend revisions before dissemination. The FBS monitors press coverage of statistics and provides clarifications in cases of erroneous interpretation of the data. As to transparency, the laws governing the collection, compilation, and dissemination of statistics are posted on the FBS website and in hard copy publications. The relevant articles are cited on the survey instruments. Internal government access to some statistics prior to their release is made known to the public on the DSBB and in the publications. Statistical outputs of the FBS are clearly identified as such and advance notice of the impending improvements to the methodology and statistical techniques is provided on the FBS website and publications. Staff behavior is guided by the rules, regulations and administrative procedures of the civil service, which are made known to the staff. Adherence to ethical standards is promoted in the training programs and assessed during the annual performance review of the staff.

27. MOF statistical policies and practices are guided by professional principles. While there are no specific laws which establish the professional independence of GFS compilers, the AGPR has autonomy under law in carrying out its accounting functions, and there is a longstanding convention of independence in MOF in compiling and publishing fiscal data. There has been no evidence of political influence in choosing either data sources or compilation methods. GFS policies and practices are broadly transparent. There are no restrictions on public access to GFS, nor is there privileged internal government access to the data. GFS are disseminated on the MOF website and in the annual MOF publication Economic Survey. Data compiled and disseminated by the MOF are identified as such, but little information is provided to users on the arrangements governing the production and dissemination of the statistics. GFS compilers are governed by the Pakistan civil service Code of Conduct, which requires them to properly apply laws and regulations and ensures that GFS policies and practices are guided by ethical principles. Staff behavior is guided by the rules, regulations and administrative procedures of the civil service, which are made known to the staff.

28. The SBP follows practices that encourage professionalism in the compilation of monetary and balance of payments statistics. SBP staff indicate that data are compiled on an impartial basis and that the choice of sources and statistical techniques is based solely on statistical considerations. The SBP provides comments on erroneous interpretation and misuse of its statistics. As for transparency, the terms and conditions under which monetary and balance of payments statistics are collected, processed, and disseminated are made available to the public. There is no internal government access to the data prior to their publication. Products of the SBP are identified as such and the SBP requests attribution when its statistics are used or reproduced. Major changes in methodology and sources are noted at the time the changes are introduced. Data suppliers for balance of payment statistics are notified in advance of major changes in methodology. The SBP Staff Regulations on staff conduct, which are made known to the staff, provide guidelines for ethical standards.

Methodological soundness

Methodological soundness refers to the application of international standards, guidelines, and agreed practices. Application of such standards, which are specific to the dataset, is indicative of the soundness of the data and fosters international comparability. Elements refer to the basic building blocks of concepts and definitions, scope, classification and sectorization, and basis for recording.

29. The national accounts follow the concepts and definitions of the System of National Accounts 1968 (1968 SNA).11 The scope of the annual accounts is very narrow and does not meet the minimum requirements for implementing the 1993 SNA. However, the FBS is instituting a comprehensive plan to revise the estimates and implement the recommendations of the 1993 SNA. It expects to disseminate the revised estimates, along with a set of quarterly accounts, during the first half of 2004. Currently, the measurement of output does not cover all own-account production and most informal activities. The classification systems are broadly consistent with internationally accepted guidelines with the exception of household consumption, which is not classified by purpose. The basis for recording is not in full accordance with the 1993 SNA. Sales taxes are excluded from the valuation of intermediate consumption. Recording is done on an accrual basis with the exception of government transactions, which are recorded on a cash basis.

30. The concepts and definitions of the CPI and WPI follow international guidelines. However, the CPI scope is limited to 52 urban centers (covering about 68 percent of all households), thus excluding households living in rural areas. The WPI does not cover small-scale manufacturing enterprises. While the national classification used for household consumption differs significantly from the international standards, the national activity classification is close to what is internationally recommended. Regarding the basis of recording, goods and services are correctly recorded at market values and on an accrual basis.

31. The concepts and definitions used in compiling GFS are broadly based on the methodology prescribed in A Manual on Government Finance Statistics, 1986 (GFSM 1986), except that the treatment of privatization proceeds resembles the methodology of the Government Finance Statistics Manual 2001 (GFSM 2001). However, the definition of government debt does not fully follow international standards because it includes government guaranteed debt. There are no specific plans to migrate to the GFSM 2001 standard. The scope of GFS is budgetary general government, which covers the bulk of general government activity, but does not cover the activity of extra-budgetary units in central government. Classification and sectorization systems only follow GFSM 1986 standards to a limited extent. Institutional sectors are not explicitly defined, since only budgetary data are covered in GFS. Central and provincial governments are distinguished, but local governments are included with the provincial governments. This is a significant problem given recent initiatives to devolve more power to the local government sector. Foreign grants received, and loans received by provincial governments from the central government, are incorrectly classified as financing, and revenue, respectively. The classification of expenditure departs substantially from GFSM 1986 methodology, both because the economic and functional classifications are not clearly distinguished, and because the level of detail falls far short of what is required. The basis of recording GFS is on, or close to, a cash basis, as required by the GFSM 1986 standard. Transactions are recorded on a gross basis. Corrective transactions are not necessarily made in the original period, as required by GFSM 1986.

32. In general, the analytical framework for monetary statistics reflects concepts and definitions that are based on the IMF’s draft Guide to Money and Banking Statistics in International Finance Statistics of December 1984. The scope of the monetary statistics is comprehensive. Classification and sectorization principles for the monetary statistics published in the Statistical Bulletin do not follow international standards. For example, positions with nonresidents are not always identified as such, securities repurchase agreements are not treated as collateralized loans, and data on financial derivatives are recorded off-balance-sheet. The basis for recording flows and stocks is largely consistent with MFSM, with accrued interest identified in the balance sheet but not included in the outstanding amount of the corresponding financial instrument. In addition to the monetary statistics that are published in the Statistical Bulletin, the SBP also prepares data for internal use and for reporting to the IMF. These data, which the SBP started publishing soon after the ROSC mission took place, use classification and sectorization systems that are broadly in line the Monetary and Financial Statistics Manual. Because these data were not publicly available at the time of the mission, they were not used as the basis for the assessment.

33. Pakistan’s balance of payments statistics are compiled broadly in accordance with the concepts and definitions of the BPM5. However, the presentation of the balance of payments on the web and in the publications of the SBP is still based on the BPM4. 12 In general, the scope of the balance of payments pertaining to residency conforms to BPM5. The exceptions include the enterprises operating in the Export Processing Zones and offshore banks, which are regarded as nonresidents, and Pakistani nationals residing abroad but owning a dwelling in Pakistan, which are considered residents. Classification and sectorization systems also follow, to some extent, BPM5. For example, the entire amount of oi1 exports under product sharing agreements is classified as direct investment in the current account; the portion of this amount relating to repatriation of direct investment capital should be recorded in the financial account. The basis for recording transactions follows best international practices.

Accuracy and reliability

Accuracy and reliability identifies features that contribute to the goal that data portray reality. Elements refer to identified features of the source data, statistical techniques, and supporting assessments and validation.

34. The source data for the national accounts are derived primarily from benchmark censuses, a regular household budget survey, and administrative sources. A comprehensive business register is updated on a continuous basis, but regular enterprise surveys are not conducted. A census of manufacturing is conducted every five years but other censuses are infrequent. In general, the censuses suffer from poor coverage of activities and very low response rates. In addition, while administrative data are received on a timely basis, delays in completing survey fieldwork hinder the use of the source data in the compilation process in a timely manner. The statistical techniques are weak. The estimates are compiled at 1980/81 prices and the reliance on benchmarks that are over five years old is widespread. Procedures used to derive the volume indicators are based on highly aggregated price indices, some of which do not reflect the underlying price changes. Procedures to adjust the data sources to improve coverage are deficient and techniques to derive exhaustive measures of GDP are not employed. Assessment and validation of source data are undertaken to determine the extent of sampling errors and non-response. However, despite the under-coverage in the surveys, non-measurement of informal activities, and weaknesses in statistical techniques, discrepancies in statistical outputs are not investigated. The FBS has conducted a comprehensive set of revision studies as part of the exercise to update the base year and implement the 1993 SNA. The studies develop experimental estimates for the major aggregates using revised data sources and statistical techniques.13

35. Source data for the CPI include the Family Budget Survey from which the weights have been derived. The weights for the WPI are based on a variety of sources, some of which are not entirely satisfactory. Both price indices use broad-based price collection surveys. Source data are timely and reasonably approximate the required definitions and accounting rules. Most statistical techniques are sound, but there are two exceptions for the CPI: (i) no adjustments are made to the results from the Family Budget Survey to resolve underreporting by households of certain types of expenditures; and (ii) the calculation of the price index for housing on the basis of smoothed construction costs does not reflect international standards. The latter problem has large repercussions for the accuracy of the CPI because of the large weight of housing. The assessment and validation of source and intermediate data and statistical outputs are adequate. Revision studies were conducted on the occasion of the change of the base year and were used to inform the statistical process.

36. Timely and reliable source data for GFS reasonably approximate the definitions, scope, classifications, valuation, and time of recording required by GFSM 1986. GFS are compiled directly from accounting or administrative data, and there is no need to apply other statistical techniques in the compilation process. Assessment and validation of source data are done mainly by vetting input data by line agencies and benchmarking preliminary data to audited final data. There are no material gaps in coverage or completeness of source data. No systematic reconciliation process is carried out to ensure that GFS is consistent with corresponding data derived from other sources. Revisions to GFS result routinely from the replacement of preliminary data by final, audited, data. No systematic studies are made of the revisions resulting from this process, but when benchmarking is done, the revisions are examined to ensure that any significant changes can be explained.

37. The main source data for monetary statistics are highly aggregated weekly statements from the Banking Supervision Department that do not provide the required breakdown of financial instruments and economic sectors. Source data are timely, with the exception of SBP data for the end of the fiscal year (end-June), which are only available with a two to three month lag 14. Statistical techniques could be improved; weekly data are entered into the database manually, which can be a source of errors. There are automated procedures to identify processing errors 15. The assessment and validation of source data are adequate, consisting of making comparisons between the banks’ weekly and monthly statements. Plausibility checks are applied to identify out-of-trend movements in intermediate data. Findings from revision studies are used internally to inform the statistical processes.

38. The main source data for balance of payments statistics are the International Transactions Reporting System (ITRS), other reports from enterprises, and administrative records of the government and the SBP. The data are broadly in line with the internationally accepted definitions, classifications, and scope and are received in a timely manner. Statistical techniques are generally sound, as evidenced by the methodology to calculate the cif/fob ratio. Procedures for assessment and validation of the quality of source data and statistical outputs include the comparison of data from different sources, for example, between data derived from the ITRS and data from government records or customs returns. Revision studies are not carried out.

Serviceability

Serviceability focuses on practical aspects of how well a dataset meets users’ needs. Elements refer to the extent to which data are relevant, produced and disseminated in a timely fashion with appropriate periodicity, are consistent internally and with other datasets, and follow a predictable revisions policy.

39. The relevance of the national accounts statistics is monitored through regular meetings of the NAC, which includes representatives of private and public sector agencies. The NAC is responsible for reviewing all aspects of the national accounts data compilation system and providing guidance to the FBS. The timeliness of the data does not meet GDDS recommendations. The data are compiled annually, thereby meeting the GDDS but not the SDDS requirements on periodicity. The national accounts statistics are consistent within the dataset, over time, and with balance of payments and government finance statistics. Revisions follow a predetermined schedule but this is not communicated to the public.

40. Monitoring of the relevance and practical utility of the CPI and WPI is undertaken through involving expert groups and occasional users’ surveys. The timeliness and periodicity of both indices are in accordance with the GDDS and the SDDS. The estimates are internally consistent in the sense that the aggregate index is invariant to the typology of the aggregation. Data on price indices are consistent over time. Revisions follow a well-established schedule; the base year is changed every ten years. Users are informed of the causes of revisions in the statistical series.

41. The fiscal tables produced as GFS output were designed to establish the basic parameters of fiscal policy and are extensively used for that purpose. MOF monitors press comments and comments arising at various economic forums, but receives little direct feedback from users of the statistics. No systematic procedures ascertain the relevance and practical utility of GFS in meeting users’ needs. The timeliness and periodicity of GFS meet GDDS recommendations, but not SDDS requirements. GFS are internally consistent, with the exception of the discrepancy between the deficit/surplus and financing data, which is larger than would be expected. The annual GFS time series appear to be consistent over time as a percentage of GDP. However, GFS are not reconcilable with balance of payments or monetary statistics. Revisions usually follow a regular and well-understood schedule, and preliminary data are clearly identified. However, corrections are made to data in the period in which an error is discovered, rather than shown as revisions to the original data. There is no published analysis or explanation of revisions.

42. The relevance of monetary statistics is largely monitored in the context of interdepartmental meetings within the SBP. A formal process of consultation with external users to solicit feedback does not exist. The analytical accounts of the SBP and the banking sector meet the timeliness and periodicity recommendations of the GDDS, with the exception of SBP data for the end of the fiscal year (June), which are only disseminated with a two-three months lag. There are large discrepancies in the positions between the SBP and scheduled banks. Consistency checks of monetary data with balance of payments statistics are done regularly, but data are not reconcilable with government finance statistics. While there is no formal revision policy, monetary data are final when first released, and revisions are infrequent.

43. The relevance of balance of payments statistics to official users is monitored through the Balance of Payments Standing Committee, but feedback on quality issues from users outside the government is not sought. Data users can address questions by electronic mail. The balance of payments data exceed GDDS recommendations and meet the SDDS requirements for periodicity and timeliness. Data are internally consistent and are verified against monetary statistics, but not with government finance statistics. Revisions follow a regular, well-established, and transparent cycle, which is made known to users. Preliminary data are identified as such.

Accessibility

Accessibility deals with the availability of information to users. Elements refer to the extent to which data and metadata are clear and readily available and to which assistance to the user is adequate to help them find and utilize the data.

44. Regarding data accessibility, national accounts are presented in a clear manner, with varying levels of detail. A brief analysis of the estimates is provided in the national accounts publication. The data are not disseminated according to a preannounced schedule though users are aware of the timeframe for publication, which has remained stable over the years. The data are presented to the National Accounts Committee before finalization and wider dissemination to the public. Metadata accessibility is generally sound. The FBS website, the DSBB, and the national accounts publication contain detailed information on the sources and methods used to compile the statistics. Assistance to users is broadly adequate, but could be improved by providing contact information for specific subjects in FBS publications and the website. A catalog of publications is available in hardcopy and on the website.

45. The data and metadata accessibility of the CPI and WPI is mostly satisfactory. The main publication presents the data in a way that facilitates proper interpretation and meaningful comparisons. However, the Family Budget Survey, which is the source of the weights for the CPI, has not been published. The dissemination media for the price indices comprise e-mails, website, and hard-copy publications. Both indices are released according to a preannounced schedule. Hardcopies of the data are released to selected users within the government as well as to the press shortly before they are published on the FBS website. Assistance to users is broadly adequate, but could be improved by providing contact information for specific subjects and publishing a catalog of FBS services.

46. GFS data have limited accessibility. The dissemination media and format are adequate, and data are released on a regular, but not preannounced, schedule. However, GFS tables reflect the particular internal needs of MOF, and do not have a clear structure. No explanations or charts accompany the data. No time series of sub-annual data is provided. Quarterly data are presented on a cumulative basis, which obscures the data for individual quarters. GFS are released to all users at the same time. There are no non-published subaggregates. GFS metadata are accessible only on the DSBB in a summary form. Assistance to users of GFS is inadequate. The contact person for GFS is publicized only on the DSBB, including telephone and facsimile numbers but not e-mail address.

47. Accessibility of monetary data needs to be improved. The SBP balance sheet, as published in the Statistical Bulletin, is not presented in an analytically useful format (e.g., no information is available on the SBP’s foreign liabilities)16. The statistics are released simultaneously to all interested parties. Nonconfidential unpublished data are made available upon request. Metadata are available on the SBP website through a hyperlink to the DSBB. Users are provided with contact information on the DSBB and the SBP website, but not in hardcopy publications. A list of publications is provided in the SBP’s Statistical Bulletin.

48. The accessibility of data and metadata on balance of payments statistics is good. Statistics and metadata are presented in a clear manner that facilitates proper interpretation and meaningful comparisons in the Statistical Bulletin and in Pakistan’s Balance of Payments. These publications are posted on the SBP website, which is also hyperlinked to the DSBB. The data are released simultaneously to all users. Nonpublished but nonconfidential data are made available upon request. Users are provided with contact information on the website but not in hardcopy publications. A list of statistical publications is included in the Statistical Bulletin and Pakistan’s Balance of Payments.

IV. Staff Recommendations

49. Based on the results of the data quality assessment, discussions with the Pakistani authorities, and responses from data users, the measures below are proposed to further enhance Pakistan’s adherence to international statistical standards. While the actions listed below should be treated with high priority, those regarding national accounts and government finance statistics need to be addressed with the greatest urgency. A more detailed list of recommendations is included in the accompanying Detailed Assessments document.

General recommendations

  • Pursue the initiative to give more autonomy to the FBS and conduct public awareness exercises aimed at improving the public perception of the FBS.

  • Undertake actions to improve response rates for surveys, for example, by amending the General Statistics Act to increase the fines for not providing information.

  • Undertake a review of statistical policies and practices in consultation with users, addressing at least the following issues: (i) presentation of statistics (tables, text, charts, etc.); (ii) development of detailed metadata for all datasets; (iii) establishment of advanced release calendars; (iv) simultaneous release of data to all users; (v) publicizing the availability of unpublished data; (v) conducting revision studies; and (vii) assisting users of official statistics by providing more detailed contact information.

  • Provide more training in statistical methodologies and modernize computer equipment.

  • Strengthen coordination among institutions to enhance intersectoral consistency of data.

  • Update and enhance regularly the GDDS metadata posted on the DSBB for all sectors.

National accounts

  • Amend the role and functions of the NAC so that the final decision on the dissemination of the statistics rests with the FBS.

  • Introduce a program of regular (annual) establishment surveys to improve the timeliness of source data. Broaden the economic activities covered by surveys.

  • Improve the timeliness of the national accounts. Disseminate the estimates that are revised for government agencies six months after the end of the reference period.

  • Implement procedures to estimate the non-observed economy. The procedures should target, in particular, informal activities that are significant in Pakistan.

Consumer and wholesale price indices

  • Implement the planned introduction of the COICOP as soon as possible.

  • Review the treatment of Housing in the CPI, preferably by replacing the technique for estimating rentals on the basis of construction costs with an index reflecting the cost of purchasing and maintaining dwellings.

  • Consider expanding the scope of the CPI to cover all urban areas, and eventually cover all Pakistan by including information on rural areas.

Government finance statistics

  • Set up a small group within the MOF whose main task is the compilation and dissemination of GFS.

  • Migrate to the GFSM 2001 classification structure, as far as possible.

  • Provide the classificatory detail specified in the GFSM 2001 and publish a time series of discrete quarterly data.

  • Examine the statistical discrepancy between the deficit/surplus (net lending/borrowing under GFSM2001) and financing for each quarter, under the supervision of the Fiscal Monitoring Committee.

Monetary statistics

  • Compile and disseminate monetary statistics based on the data collected through the monthly report forms rather than the weekly statements provided by the Banking Supervision Department.17

  • Enhance the existing data validation procedures to ensure that mutual claims between the SBP and the scheduled banks are recorded consistently.

  • Implement fully the guidelines of the MFSM on sectorization, classification, and accounting.

Balance of payments statistics

  • Compile data on the external transactions of enterprises operating in the Export Processing Zones and offshore banks and treat them as residents of Pakistan.

  • Disaggregate the counterpart of exports of crude oil under product sharing agreements into (i) an income component, to be classified in the direct investment income account, and (ii) a component for the repatriation of direct investment capital, to be classified as foreign direct investment in Pakistan in the financial account.

  • Strengthen the procedure to estimate reinvested earnings to include, in addition to earnings capitalized as bonus shares, all earnings capitalized in other reserve accounts.

  • Reclassify government-guaranteed loans from the general government sector to the sector of the beneficiary of the loan proceeds.18

  • Complete the compilation of the international investment position on the basis of BPM5.

1

The first GDP releases, which are published about two weeks before the end of the fiscal year, are based on nine to ten months’ observed data and two to three months’ projections. The first annual GDP data based on 12 month’s observations are disseminated with a lag of 11 months (vis-à-vis the recommended 6-9 months). This situation would be remedied with the publication of quarterly national accounts as presently in preparation.

2

Beginning in February 2004, biweekly data on the analytical accounts of the SBP are disseminated on the website and the Statistical Bulletin.

3

With the assistance of the authorities, an informal survey was conducted among academics, media, business associations, banks, and public sector agencies. The results of the survey are presented in Appendix IV of the accompanying document presenting the detailed assessments.

4

Beginning in June 2004, timeliness of the analytical accounts of the SBP ha been improved.

5

A detailed description of the GDDS can be found on the IMF’s Data Standards Bulletin Board (DSBB) on the Internet at http://www.dsbb.imf.org.

6

The mission team was headed by Mr. Edgar Ayales and included Messrs. Henri Hoezoo, Cornelis Gorter, Rainer Koehler, Thomas Alexander, and Ricardo Davico (all STA), and Mr. Brian Donaghue (consultant).

7

A detailed description of the SDDS can be found on the IMF’s Data Standards Bulletin Board (DSBB) on the Internet at http://www.dsbb.imf.org.

8

Beginning in June 2004, the SBP disseminates the analytical accounts based on month-end data with a lag of two weeks.

9

Information on data quality may be found in the “Data Quality Reference Site” on the IMF’s Data Standards Bulletin Board (DSBB) website (http://dsbb.imf.org/dqrs.index.htm).

10

The Generic Framework is shown in Appendix III of the accompanying document.

11

During 2004, the FBS has completed the rebasing of the national accounts at 1999-2000 prices and made progress towards the full adoption of the System of National Accounts 1993 (1993 SNA).

12

During 2004, the SBP has made progress in adopting the BPM5 format for the presentation of the data.

13

During 2004, the FBS has completed the rebasing of the national accounts at 1999-2000 prices and made progress towards the full adoption of the 1993 SNA.

14

In June 2004, the SBP started compiling monetary statistics based on month-end data.

15

Beginning in June 2004, source data are collected electronically, thereby reducing the possibilities of data-entry errors.

16

In June 2004, the SBP began disseminating monetary statistics more in line with the MFSM. Also, a publicly announced advance release calendar is available on the SBP website.

17

Beginning in June 2004, monetary statistics are based on month-end data and are more in line with the MFSM.

18

This recommendation has been adopted in 2004.

Same Series

Other IMF Content

Other Publishers

Cover IMF Staff Country Reports
Pakistan: Report on the Observance of Standards and Codes (ROSC)—Data Module
Author: International Monetary Fund
Volume/Issue:
Volume 2004: Issue 398
Publisher:
International Monetary Fund
ISBN:
9781451830613
ISSN:
1934-7685
Pages:
195
DOI:
https://doi.org/10.5089/9781451830613.002