## _cr03140 ## Source details **Canonical URL:** [_cr03140](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2003/_cr03140.pdf) ## Other formats - [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2003/_cr03140.pdf.md) - [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2003/_cr03140.pdf.json) --- ### Executive Summary — mission, scope, and overall findings - Mission: Report on the Observance of Standards and Codes (ROSC) — data dissemination module and Data Quality Assessment Framework (DQAF) application to principal macroeconomic datasets (national accounts, CPI, PPI, government finance, monetary, balance of payments). - Agencies assessed: State Department of Statistics (SDS), Ministry of Finance (MOF), National Bank of Georgia (NBG). - Mission dates: visiting Tbilisi during July 15–31, 2002. - Basis: publicly available information and information provided prior to and during the mission. - General findings: - Georgia has made good progress in improving statistics in a number of areas, but substantial shortcomings remain that could detract from accurate analysis and policy formulation. - Some shortcomings can be addressed with current resources; improving accuracy and reliability of macroeconomic statistics would require additional resources. - Over-emphasis on timeliness has compromised accuracy and/or degree of detail in some statistics. ### GDDS participation, coverage, periodicity, and timeliness - Participation status: Georgia is not a GDDS participant but largely meets GDDS recommendations for coverage of most core data categories. - Coverage exceptions: - Real sector: primary commodity and agricultural indices; unemployment. - External sector: public and publicly guaranteed external debt outstanding with maturity breakdown; public and publicly guaranteed debt service schedules. - Fiscal sector: total government debt. - Extensions encouraged by GDDS that Georgia does not meet: gross national income, capital formation and saving, government guaranteed debt, share price index, private external debt not publicly guaranteed, reserve-related liabilities, international investment position. - Periodicity and timeliness highlights (as presented in Text Table 1): - National accounts: GDDS periodicity/timeliness guidance 10–14 months; Georgia: Flash: 1 month; Prelim: 4 months; Final: 13 months. - Depository Corporations Survey: GDDS 2–3 months; Georgia: 1 month. - Balance of Payments aggregates: GDDS 6–9 months; Georgia: Q: 3 months; A: 4 months. - Consumer price index: GDDS 1 or 2 months; Georgia: 10 days. - Producer price index: GDDS 1 or 2 months; Georgia: 20 days. - Employment/wages: GDDS 6–9 months; Georgia: employment timeliness 14 months; wages/earnings 14 months. - Central government budgetary aggregates: GDDS 6–9 months; Georgia: monthly*: 2 months, annual preliminary: 2 months, annual final: 6 months. - Broad money and credit aggregates: GDDS 1–3 months; Georgia: 1 month. - Gross official reserves: GDDS 1–4 weeks; Georgia: M: 4 weeks. ### DQAF-based assessment — summary - The IMF supplemented GDDS assessment with the DQAF covering prerequisites and five dimensions of data quality: integrity, methodological soundness, accuracy and reliability, serviceability, and accessibility. - For Georgia, with the exception of balance of payments statistics, macroeconomic statistics and the statistical base are considered broadly adequate by the IMF to conduct effective surveillance. For balance of payments statistics the IMF supplements official data with staff estimates; for other macroeconomic statistics, the IMF uses official data compiled by Georgian authorities. - Several shortcomings identified that could detract from accurate analysis and policy formulation. ### Prerequisites of quality — legal, institutional, resources (key findings) - Law on Statistics (No. 1071 of 1997) provides legal framework: responsibilities, data sharing and coordination, confidentiality guarantee, and mandatory reporting authority. - Code of Administrative Offences (1999): penalties exist but are nominal and mostly applied for business registration rather than statistical compliance. - Resource constraints: - SDS staffing and financial resources inadequate for national accounts and balance of payments compilation; price statistics have adequate staff but inadequate financial and computing resources. - Computing resources generally adequate for national accounts but inadequate for balance of payments. - National accounts Division staff: nine; high staff turnover (seven left in last two years); staff salaries about US$20–30 per month. - Financing of key surveys dependent entirely on donor aid (examples: household budget survey and labor force survey). - Institutional authority: - SDS appears to lack authority over regional offices when national priorities conflict with regional priorities, hindering data collection (notably for balance of payments). - SDC (customs) not made responsible for compiling trade statistics consistent with international standards. ### Integrity — professionalism, transparency, ethics - Agencies generally demonstrate professionalism; ethical guidelines exist and staff are aware. - Transparency shortcomings: - Terms and conditions for compilation and dissemination and government/internal access prior to public release are not sufficiently publicized. - In practice selected government agencies (MOEIT, MOF, NBG, IMF) often receive data ahead of public release. - NBG specifics: - Two divisions compile separate monetary datasets (one for IMF surveillance using program exchange rates, another for IMF statistical publications using actual exchange rates), leading to unresolved methodological differences. ### Methodological soundness — dataset summaries - National accounts: - Concepts and definitions follow 1993 SNA. - GDP compiled by production and expenditure approaches annually and quarterly; volume measures available only by production approach; only production and generation of income accounts compiled from full SNA sequence. - Valuation: basic data for general government on a cash basis. - CPI: - Follows ILO guidelines and 1993 SNA; scope covers resident households in major urban areas; owner-occupied dwelling services initially not measured (later extended per Authorities’ actions). - Prices recorded on an accrual basis; weights and prices based on actual market transactions inclusive of discounts and trade and transport margins. - PPI: - Follows 1993 SNA; scope covers mining, manufacturing, electricity, gas and water; valuation at basic prices, accrual reporting. - GFS: - Annual budget laws broadly consistent with GFSM 1986 framework but with significant differences (e.g., revenue includes loan repayments; expenditure includes amortization of foreign loans). - Scope: cumulative year-to-date and annual GFS generally cover cash activities of general government, except operations in “Special Accounts”; no published data on domestic government debt. - Basis for recording: cash basis (GFSM 1986). - Monetary statistics: - Follow MFSM concepts; depository corporations survey excludes nonbank depository corporations (considered insignificant); other financial corporations (insurance, pension funds) not covered. - Balance of payments: - Follow BPM5 concepts; scope consistent except reinvested earnings are excluded; classification broadly conforms; valuation rules align with international standards except some government source data on a cash basis. ### Accuracy and reliability — principal data source and technique issues - National accounts: - Source data generally inadequate: business register not comprehensive; sector coverage uneven (industry good; transport/communications satisfactory; agriculture, retail trade, construction, catering and services poor). - Nonobserved economy estimated to be about 33 percent of GDP; significant parts of nonobserved economy not well measured. - Main weakness: techniques for compilation of value-added at constant prices do not meet international standards (single deflation rather than double deflation; no double indicator method). - Supply and use tables used but not sufficiently detailed to reconcile GDP by production and expenditure fully; statistical discrepancy for 1999 GDP about 10 percent, reduced to about 3 percent after SUT application. - CPI/PPI: - Data sources generally good; statistical techniques generally sound; indices final when first published and not revised. - Concerns: nonobserved economy may be inadequately represented in weights and price collections; PPI suffers from deficiencies in the business register. - Government finance: - Central government source data generally high quality; local government expenditure data and coverage of “Special Accounts” need improvement. - No statistical techniques used to estimate missing data; missing data not estimated. - Revisions are investigated and reasons documented. - Monetary statistics: - Data sources generally adequate; limitations in sectorization of some securities and derivatives; statistical techniques confined to automated checks; revisions rare. - Balance of payments: - Data sources inadequate; heavy reliance on administrative data (customs) that are deficient for nonobserved economy and not consistent with international trade statistics standards. - Statistical techniques, assessment, and validation largely deficient; insufficient studies of revisions between preliminary and final data. ### Serviceability — relevance, consistency, revisions, and user engagement - User consultation: ongoing formal consultation not undertaken; SDS conducted two user surveys in last two years but no advisory group systematically assessing relevance. - User survey (mission July 15–31, 2002): questionnaires sent 13; responses 11; response rate 85%; key responses: - National accounts used regularly by 8 respondents — 73%. - Price statistics used regularly by 9 respondents — 82%. - Fiscal statistics used regularly by 9 respondents — 82%. - Balance of payments used regularly by 9 respondents — 82%. - Monetary statistics used regularly by 9 respondents — 82%. - Coverage satisfactory? Yes: 4 — 36%; No: 7 — 64%. - Accuracy satisfactory? Yes: 1 — 9%; No: 10 — 91%. - Overall assessment of Georgia’s statistics: Good: 3 — 27%; Unsatisfactory: 8 — 73%. - Consistency: - Government finance and balance of payments statistics are internally consistent but largely not consistent over time. - National accounts: consistent annual time series available from 1996; earlier years not reconstructed for methodology changes. - Revision policies: - Revision policies/practices for national accounts, government finance, and balance of payments not always clear or publicized. - National accounts revisions schedule: Flash 1 month, Preliminary 4 months, Final 13 months; revisions investigated but not always documented publicly. ### Accessibility — dissemination, metadata, and user assistance - General accessibility shortcomings: - Except for monetary statistics, macroeconomic data largely not accessible due to presentation deficiencies, limited use of websites, lack of simultaneous release, and failure to publicize availability and terms for detailed unpublished statistics. - Metadata: virtually no metadata exist on sources, concepts, and methods used in compilation. - User assistance: contact information generally provided and agencies respond to user requests. - Dataset-specific accessibility notes: - National accounts: SDS website initially contained no national accounts data; preliminary/revised/final series release schedule exists but slippage may occur; unpublished nonconfidential data available on request, usually for a charge; metadata absent. - CPI/PPI: indices final at first release; SDS website initially did not include price index data; CPI released 10 days after reference month; PPI released 20 days after reference month; metadata limited to brief descriptions in SDS Yearbook (improvements planned/implemented by Authorities). - GFS: no dedicated GFS publication or website at time of mission; MOF and SDS dissemination formats broadly adequate but lack detail; nonconfidential unpublished GFS available on request; no pre-announced public release dates; internal government access prior to public release occurs. - Monetary statistics: NBG publishes monthly data on its website and quarterly MBSB; dissemination format (Acrobat/Excel files) judged "not user friendly"; metadata documentation not publicly available at mission time. - Balance of payments: SDS publications did not meet user needs; limited accompanying text/tables/analysis; no SDS website data initially (later actions planned); revisions not documented publicly. ### Authorities’ response — actions taken and planned (Response by the Authorities, May 14, 2003) - General commitments: - Make full use of official websites to facilitate simultaneous release of macroeconomic statistics to all users with specified exemptions. - Pre-release access exemptions to designated officials outside producing agencies for no more than three working days (later sector-specific statements indicate not more than two working days for some datasets). - Publish exemptions, embargo arrangements, calendar of no-later-than release dates, and availability/terms for detailed unpublished statistics. - Review State Department of Customs responsibilities for merchandise trade statistics. - Address resource adequacy within the State Program for the Development of Statistics. - SDS — National Accounts (actions taken and planned): - Actions taken: - Allocated additional staff, computer, and financial resources; secured financial resources for an agricultural census in 2004. - Introduced double deflation techniques for volume GDP measures. - Validation procedures and established revision schedule; quarterly GDP by expenditure approach disseminated since August 2002; national accounts metadata made available on the SDS web page. - Actions planned: - Propose more staff and financial resources in medium-term budget. - Post strict embargo arrangements on website; prior access not more than two working days to named agencies. - Develop volume measures of GDP by expenditure approach. - Compile value added estimates for South Osetia and Abkhasia using administrative data; improve business register exhaustiveness; undertake census surveys and broaden survey coverage. - Improve measurement of nonobserved economy and consumption of fixed capital; compile more detailed supply and use tables; develop metadata document. - SDS — CPI/PPI (actions taken and planned): - Actions taken: - Monthly CPI disseminated electronically since 2003 via SDS website; summary metadata disseminated; scope extended to include owner-occupied dwelling services; CPI weights based on COICOP. - Annual PPI disseminated electronically; summary metadata disseminated. - Actions planned: - Propose additional computer and financial resources; investigate electronic data transfer between regional and central offices. - State on website strict embargo arrangements with prior access not more than two working days. - Review quality change assessment techniques; develop concepts/sources/methods documents; disseminate monthly PPI electronically. - MOF — Government Finance Statistics (actions taken and planned): - Actions taken: - UNDP-funded project to strengthen MOF statistical capabilities: new computer equipment/software, intranet, central GFS database, MOF website. - Draft law ("Georgian Budget System") recently passed incorporating principal IMF recommendations from ROSC. - Actions planned: - Assign Macroeconomic Forecasting, Tax and Custom Policy and Financial Statistics Department as responsible for GFS compilation. - Propose additional staff/financial resources; improve legal mandate for timely local government reporting. - Bring concepts/definitions into compliance with international standards (as in draft budget law); reclassify cash data into GFS Manual 2001 classification framework (no accrual migration in medium term). - Include own revenues/expenditures from "Special Accounts" from fiscal year 2003; expand detail and reclassify nonstandard items. - Publish monthly discrete GFS in addition to cumulative year-to-date; identify preliminary and final data; publicize revision studies if undertaken. - Improve presentation and dissemination via MOF and SDS websites; publicize availability of nonconfidential unpublished data; publish aggregated metadata on concepts and definitions. - NBG — Monetary statistics (actions taken and planned): - Actions taken: - Since 2002 annual MBSB, NBG disseminates monthly data in Excel and Acrobat formats. - Actions planned: - Study concentrating monetary statistics compilation in one unit or establish procedures to resolve methodological differences. - Publish monetary data consistent with MFSM together with supplementary key data used for monitoring IMF financial program. - Collect information on nonbank depository corporations, insurance companies, and pension funds to expand coverage of depository and financial corporations surveys. - SDS — Balance of Payments (actions taken and planned): - Actions taken: - Introduced an advance release calendar for BOPS with no-later-than dates; calendar to be placed on SDS website; annual BOP data now disseminated on SDS website; BOP Division allocated two new computers. - Actions planned: - Propose BOP surveys for inclusion in State Program of Statistics under Law on Statistics mandate. - Subject to government decision, apply increased penalties for nonresponse selectively and publicize them. - Work with Customs to reconcile bilateral discrepancies if Customs made responsible for trade statistics consistent with international standards. - Introduce surveys: general BOPS survey of large enterprises; specialized inward FDI survey (priority); specialized surveys for transport, communication, construction, insurance services. - Develop techniques to estimate shuttle trade, income/expenditure of residents working abroad, and workers’ remittances. - Use adjusted partner country data for goods trade until customs coverage improves; SDS to reconcile bilateral discrepancies jointly with customs authorities in major partner countries. - Post summary and detailed quarterly and annual BOP data on SDS website and develop sources and methods documentation. ### Staff recommendations — prioritized actions (selected high priority items) - General high priority: - Participate in the IMF’s GDDS. - SDS: provide additional resources to improve source data needed to compile macroeconomic statistics. - SDS/MOF/NBG: create an advisory group representing users and producers of macroeconomic statistics. - SDS/MOF/NBG: undertake a complete review of data dissemination policies with users addressing presentation, website dissemination, analysis/commentary, advance release calendars, and publicizing availability/terms of detailed unpublished statistics. - SDS/MOF/NBG: introduce strict embargo arrangements to ensure simultaneous release or publish pre-release lists and disseminate schedule. - SDS/MOF: make SDC responsible for compiling trade statistics consistent with international standards. - SDS: significantly increase penalties for nonresponse and publicize them; improve measurement of the nonobserved economy. - National Accounts high priority: - Publish detailed annual and quarterly national accounts data for a run of years, including analysis of revisions. - Improve accuracy of source data and exhaustiveness of business register. - Strengthen statistical techniques by developing detailed supply and use tables and applying proper techniques in GDP volume compilation. - CPI/PPI high priority: - Provide additional computer and financial resources and investigate electronic data transfer between regional and central offices. - Government Finance high priority: - Identify clearly which MOF division is responsible for compiling GFS and provide adequate resources. - Formally establish MOF will compile detailed GFS fully corresponding to international standards and provide resources. - Compile discrete monthly GFS in addition to cumulative year-to-date data. - Balance of Payments high priority: - Publish detailed annual and quarterly BOPS data for a run of years, include revision analysis. - Expand basic data sources through SDS-mandated surveys (general BOPS survey of large enterprises; inward FDI survey; specialized services surveys). - Establish new techniques for shuttle trade and remittances estimation; use partner country data adjustments for trade until coverage improves. *Italicized source: IMF Mission (Statistics Department) ROSC — Data Dissemination and Data Quality Assessment (mission to Georgia, July 15–31, 2002); Response by the Authorities (May 14, 2003).* ### Executive Summary ...................................................................................................... ### Executive Summary ### Mission and scope - Assessment: Report on the Observance of Standards and Codes (ROSC) — data dissemination module and Data Quality Assessment Framework (DQAF) application to principal macroeconomic datasets (national accounts, CPI, PPI, government finance, monetary, balance of payments). - Agencies assessed: State Department of Statistics (SDS), Ministry of Finance (MOF), National Bank of Georgia (NBG). - Mission dates: visiting Tbilisi during July 15–31, 2002. - Basis: publicly available information and information provided prior to and during the mission. ### General findings - Georgia has made good progress in improving statistics in a number of areas, but substantial shortcomings remain that could detract from accurate analysis and policy formulation. - Some shortcomings can be addressed with current resources; improving accuracy and reliability of macroeconomic statistics would require additional resources. - Over-emphasis on timeliness has compromised accuracy and/or degree of detail in some statistics. ### GDDS coverage, periodicity, and timeliness - Participation status: Georgia is not a GDDS participant but largely meets GDDS recommendations for coverage of most core data categories. - Coverage exceptions: - Real sector: primary commodity and agricultural indices; unemployment. - External sector: public and publicly guaranteed external debt outstanding with maturity breakdown; public and publicly guaranteed debt service schedules. - Fiscal sector: total government debt. - Periodicity and timeliness: Georgia meets, and in many cases exceeds, GDDS recommendations for periodicity and timeliness for most core data categories. Exception: timeliness of labor market indicators (employment and wages/earnings). - Extensions encouraged by GDDS that Georgia does not meet: gross national income, capital formation and saving, government guaranteed debt, share price index, private external debt not publicly guaranteed, reserve-related liabilities, international investment position. ### GDDS quality, integrity, and access dimensions - Quality dimension: - Georgia largely does not meet GDDS recommendations on quality (limited published documentation on methodology; little component detail and reconciliation with related data). - Integrity dimension: - Legal framework generally supports integrity of statistical system, but terms and conditions for compilation and dissemination are not sufficiently publicized. - Internal government access to data prior to public release is not made public. - Access dimension: - Georgia does not meet GDDS recommendations on access: agencies do not always provide advance release calendars, data are not generally released simultaneously to all parties, and pre-release to specified officials outside producing agencies is not made public. ### Prerequisites of quality (legal, institutional, resources) - Legal and institutional framework of most institutions responsible for macroeconomic statistics is broadly adequate. - Specific shortcomings: - Balance of payments statistics: State Department of Customs (SDC) has not been made responsible for compiling trade statistics consistent with international standards. - SDS appears to lack authority to engage regional offices where there is a national need. - Staffing and financial resources are not adequate for SDS or MOF statistical work. - Most critical resource needs: national accounts and balance of payments statistics. - Quality awareness: room for improvement; accuracy/detail compromised by emphasis on timeliness. ### Integrity (professionalism, transparency, ethics) - Agencies generally demonstrate professionalism, transparency in statistical policies and practices, and provide ethical guidelines to staff. - However, terms and conditions for compilation and dissemination and government access prior to public release are not publicized. ### Methodological soundness - Concepts, definitions, scope, classification, sectorization, and basis for recording for national accounts, CPI, PPI, balance of payments, and monetary statistics broadly follow international standards. - Government finance statistics: scope and basis for recording generally sound, but concepts, definitions, and classifications largely do not follow international standards. ### Accuracy and reliability - Source data broadly accurate for: consumer price index, producer price index, government finance, monetary statistics. - Significant deficiencies in source data for: national accounts, balance of payments. - Statistical techniques: - Price, government finance, and monetary statistics: basically sound. - National accounts and balance of payments: major scope for improvement. - Shortcomings in data sources and lack of resources hinder assessment and validation of national accounts and balance of payments statistics. ### Serviceability (relevance, consistency, revisions) - Ongoing formal consultation with users on relevance is not undertaken. - User survey findings: - Disseminated data broadly meet user needs for monetary and price statistics. - Less adequate for national accounts and balance of payments statistics. - For government finance statistics, users' needs were assessed only for budget execution data (broadly met), not for the full GFS framework used by mission. - Internal consistency: - Government finance and balance of payments statistics are internally consistent. - Largely not consistent over time. - Unobserved economy: deficiencies affect all macroeconomic statistics; unobserved economy may comprise a third or more of GDP. - Timeliness and periodicity often exceed GDDS recommendations, but achievement has come at expense of other quality aspects. - Revision policies and practices for national accounts, government finance, and balance of payments are not always clear. ### Accessibility (presentation, dissemination, metadata) - Except for monetary statistics, macroeconomic data are largely not accessible to users due to: - Deficiencies in presentation of statistical information. - Limited use of websites for dissemination of official statistics. - Lack of simultaneous release to all users and absence of publicized embargo/prerelease arrangements. - Failure to publicize availability and terms for detailed unpublished statistics. - Metadata: virtually no metadata exist on sources, concepts, and methods used in compilation. - User assistance: contact information is generally provided and agencies respond to user requests. ### Plans for improvement (projects with secured or expected funding) - SDS: - Proposed amendments to Code of Administrative Offences to increase penalties for non-, late, or incorrect responses to surveys. - National accounts: plan to collect discrete (not cumulative) data for quarterly estimates; undertake labor force surveys and a Census of Agriculture (mostly donor dependent); provide more computing resources to regional/local offices; plan to disseminate national accounts via website. - Prices: plan to disseminate CPI/PPI via website and improve presentation in publications. - Balance of payments: plan to complete new quarterly inward FDI survey, create an international investment position statement, revise migration-related data using latest population census, and disseminate BOP data on website. - MOF: - UNDP-funded project to strengthen MOF statistical/data compilation capabilities including new computer equipment/software, intranet for information flow, creation of a central GFS database, and establishment of a MOF website. - NBG: - Strengthen adoption of the Monetary and Financial Statistics Manual (MFSM), improve sectorization and classification for NBG and other depository corporations. - Expand coverage of depository corporations survey to include nonbank depository corporations; include insurance companies and pension funds to build a financial corporations survey in the longer term. ### Comparative tabulated timeliness/periodicity highlights (as presented in Text Table 1) - National accounts: GDDS periodicity/timeliness guidance 10–14 months; Georgia: Flash: 1 month; Prelim: 4 months; Final: 13 months. - Depository Corporations Survey: GDDS 2–3 months; Georgia: 1 month. - Balance of Payments aggregates: GDDS 6–9 months; Georgia: Q: 3 months; A: 4 months. - Consumer price index: GDDS 1 or 2 months; Georgia: 10 days. - Producer price index: GDDS 1 or 2 months; Georgia: 20 days. - Employment/wages: GDDS 6–9 months; Georgia: employment timeliness 14 months; wages/earnings 14 months. - Central government budgetary aggregates: GDDS 6–9 months; Georgia: monthly*: 2 months, annual preliminary: 2 months, annual final: 6 months. - Broad money and credit aggregates: GDDS 1–3 months; Georgia: 1 month. - Gross official reserves: GDDS 1–4 weeks; Georgia: M: 4 weeks. *Italicized entries and table specifics are presented in the source Text Table 1. *Source: IMF Mission (Statistics Department) ROSC — Data Dissemination and Data Quality Assessment (mission to Georgia, July 15–31, 2002).* ### 15. Since mid-2001, the IMF complements the GDDS elements of the ROSC data module ### _cr03140 - 15. Since mid-2001, the IMF complements the GDDS elements of the ROSC data module ### Overview - Since mid-2001, the IMF complements the GDDS elements of the ROSC data module with an assessment of data quality based on the IMF’s DQAF. The DQAF comprises a generic framework and a set of dataset-specific frameworks.3 The frameworks cover a set of prerequisites and five dimensions of data quality—integrity, methodological soundness, accuracy and reliability, serviceability, and accessibility.4 - An assessment was conducted for six macroeconomic datasets: national accounts, consumer price index, producer price index, government finance, monetary, and balance of payments statistics, using the dataset-specific DQAF frameworks. - For Georgia, with the exception of balance of payments statistics, macroeconomic statistics and the statistical base are considered broadly adequate by the IMF to conduct effective surveillance. For balance of payments statistics the IMF supplements official data with staff estimates; for other macroeconomic statistics, the IMF uses official data compiled by Georgian authorities. - Several shortcomings in statistical practices were identified that could detract from accurate analysis and policy formulation. *Findings are presented following the DQAF structure and include standardized qualitative assessments using a four-part scale (See Table 2).* ### Prerequisites of Quality - Legal and institutional environment: - The Law on Statistics (No. 1071 of 1997) defines responsibilities, provides for data sharing and coordination, calls for adherence to appropriate statistical methodology and international standards by other government agencies, guarantees confidentiality of individual responses, and authorizes mandatory statistical reporting. - Under the Code of Administrative Offences (1999), penalties may be assessed for non, late, or incorrect response. Financial penalties for late or incorrect responses are nominal and have mostly been applied to ensure business registration rather than completion of statistical returns. - Law on Budget Systems and Budgetary Powers, as amended (No. 249-IIs of 1999), and the Organic Law on the NBG (1995) assign responsibility for GFS to the MOF and monetary statistics to the NBG, while the SDS retains responsibility for dissemination of all official statistics. - Potential conflicts: possible conflict between the Law on Statistics and the Anti-Monopoly Law of Georgia; confidentiality may be compromised by absence of aggregation rules in SDS databases. - Institutional authority and coordination: - SDS appears to lack authority over regional offices when national priorities conflict with regional priorities (notably for balance of payments statistics), handicapping collection. - To ensure completion of statistical returns, SDS relies on agreement with MOF requiring SDS district offices to verify SDS returns before MOF accepts tax returns; nonetheless nonresponse rates for small and medium establishments are quite high. - Resources and quality awareness: - SDS staffing and financial resources are inadequate for national accounts and balance of payments compilation; price statistics have adequate staff but inadequate financial and computing resources; computing resources are generally adequate for national accounts but inadequate for balance of payments. - As a result of resource constraints, significant parts of the work program are financed from donor sources and reflect donor priorities; national accounts and balance of payments have suffered from switching resources, high staff turnover, and low salaries affecting motivation. - No unit is charged with oversight or guidance on quality issues; resource constraints limit staff initiatives to assess quality (e.g., development of new data sources and techniques). - Timeliness requirements have been relaxed in some areas, but accuracy remains compromised by over-emphasis on timeliness. - MOF-specific: - Under Law on Budget Systems and Budgetary Powers and related instruments, MOF has adequate legal/institutional environment for GFS collection/processing/dissemination primarily for budget monitoring/management; in practice SDS disseminates aggregated GFS while MOF provides more detailed GFS to government and specified users. - Post-restructuring, responsibility for GFS compilation is not clearly defined and appears assigned to two departments. - Timely compliance by local governments in reporting to MOF is a problem (arrangements not supported by specific laws). - Financial resources for GFS compilation are largely inadequate; recruitment and retention of qualified staff is a problem; computing resources are generally adequate. - Quality awareness is generally good, but frequent organizational changes and short reporting notice may compromise data quality and control; tradeoffs are being made between timeliness and degree of detail. - NBG-specific: - Under the Organic Law of the NBG (1995), the NBG has an adequate legal/institutional environment for monetary statistics. - NBG disseminates monthly data immediately on its website and subsequently in its quarterly Monetary and Banking Statistics Bulletin (MBSB). - Two divisions compile separate sets of monetary statistics: one for dissemination and reporting to IMF for surveillance (using exchange rates set for the financial program) and another for inclusion in IMF statistical publications (using actual exchange rates). This arrangement does not encourage resolution of methodological differences. - Resources are adequate; salaries are competitive. NBG exercises quality control at every stage. No formal user surveys are undertaken. ### Integrity - SDS: - Law on Statistics positions SDS as an independent, impartial data-producing agency. Choices of sources and methods are based on professional statistical considerations; staff are free from political influences. - SDS website provides terms and conditions for data handling and confidentiality; staff are encouraged to comment on media misinterpretation. - Transparency shortcomings: dissemination practices are not publicized, resulting in selected government agencies getting advance access to national accounts, balance of payments, and price statistics before other users. Advance notice is generally given about major changes in national accounts and prices (but not balance of payments). Press conferences accompany major releases. - Ethical standards for civil servants are laid down and staff are aware. - MOF: - Professionalism in GFS compilation is constrained by budget law requirements; GFS are compiled primarily for government budget monitoring and may not fully meet international user needs. - Transparency: terms and conditions for compilation/dissemination (including internal government access prior to release) are not publicized. - Ethical standards for MOF staff are in the civil service code of conduct and staff are aware. - NBG: - Statutory provisions support independence in compilation. Staff are free from political influence and encouraged to comment on media misinterpretations. - NBG Organic Law (available on its website) contains terms and conditions for data handling. Government agencies and senior NBG management have access to data prior to release; this access is not publicly identified. Changes in methodology/source data are announced simultaneously with releases; advance notice is given only for major changes. - Ethical standards for NBG staff are codified and staff are aware. ### Methodological Soundness - Territorial coverage: - Macroeconomic statistics treat Abkhasia and South Ossetia as part of Georgia’s economic territory; in practice coverage largely excludes transactions in these regions because of lack of data. - National accounts: - Concepts and definitions follow the System of National Accounts 1993 (1993 SNA). - GDP estimates by production and expenditure are compiled and disseminated annually and quarterly. Volume measures of GDP are available only by the production approach. - From the full sequence of accounts in the 1993 SNA, only the production and generation of income accounts are compiled and disseminated. - Classification and sectorization conform to international standards. - Valuation rules align with international standards except basic data for general government are on a cash basis. - Consumer Price Index (CPI): - CPI follows ILO guidelines for CPIs and 1993 SNA for household consumption. - Scope: broadly consistent with international standards—covers resident households in major urban areas and families of all sizes and income levels; excludes households in other urban areas and rural areas; owner-occupied dwelling services are not measured. - Classification of consumption activity follows international standards. - Basis for recording: weights and prices based on actual market transactions inclusive of discounts and trade and transport margins; prices recorded on an accrual basis. - Producer Price Index (PPI): - PPI follows 1993 SNA concepts and definitions. - Scope covers mining, manufacturing, electricity, gas and water sectors. - Classification and sectorization consistent with international standards. - Basis for recording is consistent with international standards: transaction prices, valued at basic prices, accrual reporting. - Government Finance Statistics (GFS): - Annual budget laws are generally consistent with GFSM 1986 analytic framework, though concepts/definitions differ in significant respects: revenue includes repayments of loans previously extended by government; expenditure includes amortization of foreign loans; foreign financing defined on a gross basis. - Scope: cumulative year-to-date and annual GFS generally cover all cash activities of general government, except for general government spending agencies’ operations relating to own revenues collected in their "Special Accounts."5 No data on domestic, and thus total, government debt are disseminated. - Classification and sectorization broadly consistent with international standards, except economic classification of expenditure and lending minus repayments includes significant misclassifications and nonstandard items. - Basis for recording: cash payments and receipts (in accordance with GFSM 1986). Debt liabilities recorded at nominal values (GFSM 1986) rather than market values (GFSM 2001). - Monetary statistics: - Follow concepts and definitions of MFSM. - Scope generally consistent with international standards, though nonbank depository corporations are not covered in the depository corporation survey (considered insignificant), and other financial corporations (insurance companies, pension funds) are not covered in the financial corporations survey. - Classification/sectorization generally consistent, except transactions are not derived from stock data and accrued interest is only partly sectorized. - Basis for recording generally consistent with international standards. - Balance of payments: - Follow concepts/definitions of Balance of Payments Manual, fifth edition. - Scope consistent with international standards, except reinvested earnings are excluded. - Classification/sectorization broadly in conformity. - Basis for recording: valuation rules align with international standards, except source data for general government are on a cash basis. ### Accuracy and Reliability - National accounts data sources: - Data sources available for national accounts compilation are generally inadequate. - Business register coverage is not comprehensive. - Coverage by sector (in terms of value-added): industry good; transport and communications satisfactory; agriculture, retail trade, construction, catering and services poor. - With the exception of retail trade and catering, sample surveys are not used. - Administrative sources are used to estimate the nonobserved economy (unregistered enterprises, unrecorded activity by registered enterprises, unincorporated enterprises); these data sources are poor. - Data for imports and exports of services (from balance of payments) are poor. - Lack of census data undermines establishment and enterprise surveys and removes benchmarks, though available data approximate required definitions, classifications, valuation, and time of recording.6 - Statistical techniques and validation: - Main weakness: techniques used for compilation of value-added at constant prices do not meet international standards. - For enterprise and establishment surveys, districts do not undertake assessment and validation of source data; central office undertakes only limited efforts. - Assessments of discrepancies of major intermediate data are undertaken, but supply and use tables are not sufficiently detailed to fully reconcile GDP by production and expenditure approaches. - Revision studies are undertaken but not documented. *Italic source attribution: IMF staff assessment as presented in the ROSC data module excerpt.* ### 34. The data sources available for the compilation of the consumer and producer price ### _cr03140 - 34. The data sources available for the compilation of the consumer and producer price ### Data sources and methodological assessment - Consumer and producer price indices - Data sources: generally good. - Concern: nonobserved economy may be inadequately represented in weights and price collections. - Producer price index: suffers from deficiencies in the business register. - Statistical techniques: generally sound for both indices, but approach to assessing quality changes needs improvement. - Assessment and validation of source and intermediate data: follow international standards. - Revision practice: indices are final when first published and are not revised. - Government finance statistics (GFS) - Data sources: generally adequate. - Scope for improvement: coverage of general government units’ transactions (e.g., degree of detail and activities in “Special Accounts”); significant scope for improvement in local governments’ expenditure source data (degree of detail and classification structure). - Central government source data: generally high quality, regularly assessed and validated. - Constraints: lack of resources and/or legal constraints between central and local levels mean no systematic efforts to address source data gaps. - Use of statistical techniques: not needed because sources are generally comprehensive; missing data are not estimated. - Revisions: always investigated and reasons are documented. - Monetary statistics - Data sources: generally adequate. - Limitation: do not provide separate data for financial derivatives and the sectorization of securities (currently not significant). - Statistical techniques: confined to automated checks in the NBG monetary database to identify processing errors; use is adequate. - Assessment and validation: checking reported balance sheet data against secondary sources; adequate. - Revisions: rare, always investigated; reasons for revisions are not always documented. - Balance of payments statistics (BOPS) - Data sources: inadequate. - SDS practices: does not conduct BOPS surveys (except FDI survey with poor response, and a transportation services survey); relies mainly on administrative data collected by other agencies. - Customs data for merchandise trade: especially deficient for the nonobserved economy and not in accordance with international standards for trade statistics. - Statistical techniques: largely deficient. - Assessment and validation: largely deficient. - Revisions: insufficient studies of direction and magnitude of revisions between preliminary and final data. ### Serviceability — relevance, timeliness, periodicity, consistency, revisions - Relevance and user consultation - SDS monitors relevance of official statistics to government needs primarily through consultations with government agencies and some academic institutions. - SDS has conducted two user surveys in the last two years aimed at a wider audience, including subscribers. - No advisory group systematically assesses relevance; no established process to review how well user needs are met. - National accounts - Timeliness and periodicity: exceed the GDDS recommendations. - Consistent annual time series: available from 1996; earlier years not reconstructed for methodology changes. - Revision policy/practice: revisions by activity follow an established schedule that is not publicized and not generally known. Preliminary data clearly identified. - Consumer and producer price indices - Timeliness and periodicity: meet the GDDS recommendations. - Consistency: indices are consistent internally and over time. - Revision policy/practice: indices are final when first published and are not revised; policy well-known to the public. - Government finance statistics - MOF: no regular formal consultation with users; feedback is ad hoc through individual requests. - Methodological basis and aggregated classification structure: not always sufficient for analytic purposes. - Timeliness and periodicity: exceed the GDDS recommendations. - Internal consistency: data are internally consistent; inconsistencies over time can be ascribed to changes in budget laws. - GFS time series: consistent series feasible but not compiled. - Revision policy/practice: not always clear — revisions to historical monthly series are not visible because only cumulative year-to-date data are disseminated; preliminary data not always clearly identified. Reasons for revisions are investigated but only made public in response to individual user enquiries. - Monetary statistics - NBG: no formal consultation process but statistics generally considered relevant. - Timeliness and periodicity: exceed the GDDS recommendations. - Consistency: data consistent internally and over time; checks with BOPS conducted regularly and with GFS occasionally. - Revisions policy: none; monetary data considered final. - Balance of payments statistics - SDS lack of response to user needs led NBG to develop its own estimates. - Periodicity: meets GDDS; timeliness exceeds GDDS. - Consistency: substantial movements in year-to-year errors and omissions indicate data are not consistent over time. - Revision policy/practice: no attempt to document or explain revisions to the public; preliminary data not identified. ### Accessibility — dissemination, metadata, user assistance - National accounts - Presentation in SDS publications does not meet user needs for data accessibility: insufficient detail; publications too selective; difficult to assemble complete accounts over time. - SDS website: no national accounts data included. - Release practice: preliminary and annual data released according to a pre-announced schedule published each quarter, but in practice slippage may occur; government agencies often receive data a day or two ahead. - Nonconfidential unpublished data: made available on request, usually for a charge; availability and terms not publicized. - Metadata accessibility: no publication on sources and methods; available summary textual material insufficient. - Assistance: contact persons included in statistical publications; list of publications in SDS website catalogue. - Consumer and producer price indices - Data accessibility: not adequate. - Publications: contain clear tables and charts of detailed time series but no analysis of current period changes. - SDS website: no price index data included. - Release schedule: pre-announced but dates not always met; no set embargo times; government agencies receive data ahead of others. - Nonconfidential unpublished index numbers: available on request; availability and terms not publicized. - Metadata: no sources and methods publication; only inadequate descriptions in the SDS Yearbook. - Assistance: contact information included in statistical publications and catalogue. - Government finance statistics - Presentation by SDS and NBG largely does not meet user needs: lack of detail, methodological variations, and lack of a GFS time series hinder interpretation and comparisons. - Dissemination: no dedicated GFS publication or website; SDS and NBG media broadly adequate; nonconfidential unpublished GFS available upon request but terms not publicized. - Release schedule: no pre-announced public release dates; government users receive GFS before general release. - Metadata: only annual budget laws and the Law on Budget Systems and Powers, as amended, exist on concepts, sources, and methods; no GFS contact information in dissemination media. - Monetary statistics - Presentation in NBG publications generally meets user needs. - Monthly data: published electronically on the NBG website. - Dissemination format: not user friendly. - Release schedule: no pre-announced schedule; government agencies receive data ahead of general release. - Nonconfidential unpublished data: available on request; availability and terms not publicized. - Metadata: no sources and methods information included in NBG publications. - Assistance: contact information provided on the NBG website and in the quarterly MBSB. - Balance of payments statistics - SDS publications do not meet user needs for accessibility: summary quarterly and annual data released without accompanying text, tables, or charts; no analysis; no data on SDS website. - Release schedule: no pre-announced schedule; government agencies receive data ahead of general release. - Metadata: no sources and methods information in SDS publications. - Assistance: specific contact information is given for users. ### Staff recommendations — priorities and actions - General recommendations — High priority - Participate in the IMF’s GDDS. - SDS: Provide additional resources to improve source data needed to compile macroeconomic statistics. - SDS/MOF/NBG: Create an advisory group representing users and producers of macroeconomic statistics to ensure relevance. - SDS/MOF/NBG: Undertake a complete review of data dissemination policies with users addressing at least: - (i) presentation of statistical information (tables, text, charts, etc.); - (ii) greater use of websites for dissemination of official statistics; - (iii) analysis of statistics (main features, commentary); - (iv) advance release calendars indicating, at least, no-later-than release dates; - (v) publicizing availability (including terms and conditions) of detailed unpublished statistics. - SDS/MOF/NBG: Introduce strict and unambiguous embargo arrangements to ensure simultaneous release of data, or prerelease to a list of persons or officials holding designated positions within the government, but outside the producing agency (such as for two working days), and disseminate the schedule. - SDS/MOF: Make revisions policies and practices known to the public. - SDS/MOF: Make the SDC responsible for compiling trade statistics consistent with international standards for coverage and valuation of goods crossing the customs frontier. - SDS: Significantly increase penalties for nonresponse to statistical surveys under the State Program of Statistics, apply penalties selectively, and publicize them. - SDS: Improve efforts to measure the nonobserved economy, particularly regarding sources of income and of expenditure. - General recommendations — Other key - Enhance accessibility of all macroeconomic statistics by developing detailed metadata on sources, concepts, and methods, and publicize their availability. - National Accounts — High priority - Publish detailed annual and quarterly national accounts data for a run of years, including an analysis of revisions. - Improve accuracy of national accounts source data and exhaustiveness of the business register. - Strengthen statistical techniques by developing detailed supply and use tables and applying proper techniques in GDP volume compilation. - National Accounts — Other key - Extend scope to a full sequence of national accounts covering production, income, and capital accounts by institutional sector. - Compile volume measures of GDP by the expenditure approach (requires development of a system of price indexes for deflation). - Consumer and Producer Price Indices — High priority - Provide additional computer and financial resources and investigate means of electronic data transfer between regional offices and central office. - Consumer and Producer Price Indices — Other key - Review techniques used for applying quality adjustments to product prices. - Expand the producer price index to include output of the agricultural sector. - Government Finance Statistics — High priority - Identify clearly which MOF division is responsible for compiling GFS and provide adequate resources. - Formally establish that the MOF will compile detailed GFS fully corresponding to international standards in addition to existing GFS that correspond to budget laws; provide resources to facilitate this. - Clarify SDS and MOF responsibilities for GFS dissemination along lines followed by NBG for monetary statistics. - For internationally consistent GFS, ensure concepts, definitions, classifications, and basis for recording revenue and grants, expenditure and lending minus repayments, and financing are defined accordingly. - Expand coverage of GFS, at least annually, to include activities of government agencies’ own revenue in “Special Accounts.” - Compile discrete monthly GFS in addition to existing cumulative year-to-date data. - Government Finance Statistics — Other key - Compile and disseminate a consistent annual GFS time series (e.g., five years) with explanations for inconsistencies and breaks (e.g., due to changes in budget laws). - Monetary Statistics — Key recommendations - Publish monetary data consistent with the MFSM (as reported by NBG to the IMF’s Statistics Department) together with a supplementary set of key data used for monitoring the financial program with the IMF. - Collect information on nonbank depository corporations to expand the depository corporations survey coverage; complete collection of information additional to chart of accounts for commercial banks to comply with MFSM sectoral classification of financial instruments. - Collect information on insurance companies and pension funds to expand financial corporations survey coverage. - Concentrate compilation work of monetary statistics in one unit within the NBG. - Balance of Payments Statistics — High priority - Publish detailed annual and quarterly BOPS data for a run of years, including an analysis of revisions. - Expand basic data sources through introduction of statistical surveys conducted with the full legal mandate of the SDS, for example: - Conduct a general BOPS survey covering all large enterprises to collect information on major nonmerchandise BOPS transactions (services credits and debits, investment income, trade credits, loans, and other financing). - Conduct a specialized survey for inward foreign direct investment. - Conduct specialized surveys for transport, communication, construction, and insurance services credits. - Establish new statistical techniques to estimate shuttle trade, income and expenditure of residents working abroad, and workers’ remittances. - Until trade statistics coverage improves, use adjusted partner country data for imports and exports of goods; SDS and SDC should reconcile bilateral discrepancies jointly with customs authorities in major partner countries. *Italicized source: Staff Report content unit _cr03140 - 34 from the provided PDF excerpt.* ### 0. Prere ### 0. Prere uisites of Qualit y ### 0. Prerequisites overview - 0.1 Legal and institutional environment: LO LO LO LO LO LNO - 0.2 Resources: LNO LO LO LNO O LNO - 0.3 Quality awareness: LO LO LO LO LO LO - SDS: 18, 19 MOF: 20 NBG: 21 ### 1. Integrity - 1.1 Professionalism: O O O LO O O - 1.2 Transparency: LNO LNO LNO LNO LNO LNO - 1.3 Ethical standards: O O O O O O - SDS: 22 MOF: 23 NBG: 24 ### 2. Methodological soundness - 2.1 Concepts and definitions: O O O LNO O O - 2.2 Scope: LO LO O LO LO LO - 2.3 Classification/Sectorization: O O O LNO LO LO - 2.4 Basis for recording: LO O O O O LO - NAS: 26 CPI: 27 PPI: 28 GFS: 29 MFS: 30 BOP: 31 ### 3. Accuracy and Reliability - 3.1 Source data: LNO LO LO LO O LNO - 3.2 Statistical techniques: LNO LO LO O O LNO - 3.3 Assessment and validation of source data: LNO O O LO O LNO - 3.4 Assessment and validation of intermediate data and statistical outputs: LO O O O O LNO - 3.5 Revision studies: LO O O O LO LNO - NAS: 32, 33 CPI/PPI: 34 GFS: 35 MFS: 36 BOP: 37 ### 4. Serviceability - 4.1 Relevance: LO LO LO LNO LO LNO - 4.2 Timeliness and periodicity: O O O O O O - 4.3 Consistency: LO O O LO O LNO - 4.4 Revision policy and practice: LO O O LO O LNO - NAS: 38, 39 CPI/PPI: 38, 40 GFS: 41 MFS: 42 BOP: 38, 43 ### 5. Accessibility - 5.1 Data accessibility: LNO LNO LNO LNO LO LNO - 5.2 Metadata accessibility: LNO LNO LNO NO NO LNO *Source: _cr03140 - 0. Prere* ### 5.3 Assistance to users ### _cr03140 - 5.3 Assistance to users ### Overview - Document: Response by the Authorities to the IMF Data Module ROSC (May 14, 2003). - Government intention: make full use of official websites to facilitate simultaneous release of macroeconomic statistics to all users subject to specified exemptions. - Pre-release access statements: - Overview: exemptions will name designated officials outside the data producing agency that will have pre-release access to the data for a period of no more than three working days. - Later sector-specific statements (SDS, MOF statements) indicate prior access for a period of not more than two working days for some datasets. - Public web disclosures to include: - Exemptions and embargo arrangements. - A calendar of no later than release dates for provisional, revised, and final macroeconomic statistics. - Availability and terms/conditions for detailed unpublished statistics on request. - Government commitment: review State Department of Customs responsibilities for merchandise trade statistics to meet needs of both Customs and the State Department of Statistics (SDS). - Resource adequacy to be addressed within the State Program for the Development of Statistics. ### I. National Accounts Statistics — Actions Taken - SDS allocated additional staff, computer, and financial resources for national accounts compilation; high staff turnover and low salaries remain a problem. - Financial resources secured for an agricultural census in 2004. - Improvements implemented: - Improved accuracy of household income and expenditure data. - Introduction of double deflation techniques for volume measures of GDP. - Validation procedures for assessment of source data and intermediate data. - Established a revision schedule for annual and quarterly national accounts data; users made aware of revision policy and practice. - Since August 2002, quarterly GDP estimates by the expenditure approach disseminated. - Quarterly national accounts by production and expenditure approaches published in the SDS monthly bulletin. - Preliminary, revised, and final annual national accounts released on newly created SDS website with improved presentation (sequence of accounts, constant price estimates, deflators of GDP, shares, charts, and tables). - National accounts metadata made available on the SDS web page. ### I. National Accounts Statistics — Actions Planned - Prerequisites of Quality — resources: - SDS will propose more staff and financial resources within the medium-term budget for national accounts and basic economic surveys. - Integrity — transparency: - SDS will post on its website that strict and unambiguous embargo arrangements apply to preliminary, revised and final national accounts data; prior access of not more than two working days to {name agencies outside the SDS}; users requested to attribute data to the SDS. - Methodological soundness — scope: - Develop volume measures of GDP by expenditure approach. - Accuracy and Reliability — source data: - Compile value added estimates for South Osetia and Abkhasia based on administrative data. - Improve exhaustiveness of business register; undertake census surveys to improve benchmarking; broaden enterprise and household survey coverage. - Accuracy and Reliability — statistical techniques: - Improve constant price estimates, measurement of the non-observed economy, and estimation of consumption of fixed capital. - Assessment and validation: - Continue improved procedures for coverage, sampling and non-sampling errors, and response rates. - Compile more detailed supply and use tables to allocate GDP discrepancy between production and expenditure approaches. - Accessibility — metadata: - Develop a document outlining concepts, definitions, classifications, data sources, and statistical techniques used for national accounts. ### II. Consumer Price Index — Actions Taken - Monthly CPI disseminated electronically since 2003 via SDS website. - Summary metadata on concepts, definitions, classifications, data sources and techniques disseminated. - Scope extended to include owner occupied dwelling services. - CPI weights based on COICOP classification of expenditures. ### II. Consumer Price Index — Actions Planned - Legal and institutional framework — resources: - SDS will propose additional computer and financial resources in the medium-term budget and investigate electronic data transfer between regional offices and central office. - Integrity — transparency: - SDS will state on its website that strict embargo arrangements apply to CPI to ensure simultaneous release; prior access for not more than two working days to {name agencies outside the SDS}; users requested to attribute data to the SDS. - Accuracy and reliability — statistical techniques: - Review quality change assessment techniques used for applying quality adjustments to product prices. - Accessibility — metadata: - Develop a concepts, sources, and methods document or other descriptive material about the CPI. ### III. Producer Price Index — Actions Taken - Summary metadata on concepts, definitions, classifications, data sources and techniques disseminated. - Annual PPI disseminated electronically. ### III. Producer Price Index — Actions Planned - Legal and institutional framework — resources: - SDS will propose additional computer and financial resources within the medium-term budget and investigate electronic data transfer between regional and central offices. - Integrity — transparency: - SDS will state on its web page that strict embargo arrangements apply to PPI; prior access for not more than two working days to {name agencies outside the SDS}; users requested to attribute data to the SDS. - Accuracy and reliability — statistical techniques: - Review quality change assessment techniques for quality adjustments to product prices. - Accessibility — data accessibility: - Improve data accessibility via SDS web page and disseminate the monthly index electronically. - Accessibility — metadata: - Develop a concepts, sources, and methods document for the PPI and include metadata on the SDS web page. ### IV. Government Finance Statistics — Actions Taken - MOF UNDP-funded project to strengthen statistical/data compilation capabilities, including: - Obtaining new computer equipment and software. - Developing an intranet to ensure information flow. - Creating a central GFS database. - Establishing a GFS website. - MOF created its own website; planned content referenced in actions planned. - Draft law ("Georgian Budget System") recently passed in Parliament incorporating principal IMF recommendations from ROSC—Data Module. ### IV. Government Finance Statistics — Actions Planned - Prerequisites — legal/institutional: - Macroeconomic Forecasting, Tax and Custom Policy and Financial Statistics Department in MOF responsible for GFS compilation. - MOF will study ways to improve legal mandate/measures to ensure timely reporting of detailed data by local governments. - Resources: - MOF will propose additional staff and financial resources for GFS compilation within medium-term budget. - Merging Ministry of Revenue and Ministry of Finance may free resources. - No additional resources will be made available to migrate to GFS Manual 2001. - Quality awareness: - Document quality control processes; communicate trade-offs between quality, timeliness, and degree of detail (to be addressed in UNDP project). - Integrity — professionalism: - MOF recognizes need for detailed GFS corresponding to international standards (budget law to reflect this). - Integrity — transparency: - MOF will state on website/publications terms and conditions of GFS compilation, approval processes, that there are no restrictions on access to GFS, and that internal government access occurs before publication. - MOF (with SDS and NBG) will identify MOF as GFS compiler in all GFS publications. - Methodological soundness: - Bring revenue, expenditure, and financing concepts and definitions into compliance with international standards (including definition of expenditure) as included in draft budget law. - Reclassify existing cash data into GFS Manual 2001 classification framework; no move to accrual accounting in medium term. - Scope and classification: - From fiscal year 2003, own revenues and expenditures from own revenue sources included. - Expand detail for all GFS tables, eliminate non-standard items, and reclassify certain current and capital expenditures and amortization of foreign financing to appropriate GFS categories. - Source data: - New budget law reforms for State Treasury Department to improve recording of revenues, expenditures, financing: - Improve details of extrabudgetary funds’ data for full consolidation and analysis. - Improve detail and classifications of local government source expenditure classifications to comply with international standards. - Improve availability of subsidies and other current transfers to compile complete economic classification of expenditures. - Serviceability — relevance and consistency: - MOF will solicit formal feedback from users via website publication of GFS. - Due to lack of resources, MOF does not plan to compile a consistent GFS time series in the near term; will provide explanations of inconsistencies and non-comparability when disseminating data. - Revision policy and practice: - MOF plans to publish monthly data in addition to cumulative year-to-date data; identify preliminary and final data. - Will state policy to make public results of studies/analysis of data revisions and post studies on website if undertaken. - Accessibility — data and metadata: - Improve presentation of aggregated GFS to conform with international standards; include monthly data and recent historical data. - Release GFS on a pre-announced schedule via MOF and SDS websites; make GFS available to all users outside central government simultaneously. - Publicize availability of non-confidential, nonpublished data upon request. - Due to lack of resources, MOF does not plan to compile detailed metadata in medium term but will compile aggregated metadata on concepts and definitions. - Assistance to users: - MOF will publicize contact persons for various GFS subject fields on its website and in hard copy publications. ### V. Monetary Statistics — Actions Taken - Starting with the annual bulletin of monetary and banking statistics of 2002, the NBG disseminates monthly data in Excel files and Acrobat format. ### V. Monetary Statistics — Actions Planned - Prerequisites — legal/institutional environment: - NBG will study possibility of concentrating monetary statistics compilation in one unit or establish procedures to resolve methodological differences between two currently produced data sets. - Methodological soundness — concepts and definitions: - NBG will publish monetary data consistent with the MFSM (as reported by NBG to the IMF’s Statistics Department) together with a supplementary set of key data used for monitoring the financial program with the IMF, if needed. - Scope: - NBG will collect information on quantitatively significant nonbank depository corporations to expand coverage of depository corporations survey and complete collection of information additional to the chart of accounts for commercial banks to comply with sectoral classification of financial instruments recommended by the MFSM. - In the medium term, NBG will collect information on insurance companies and pension funds to expand coverage of the financial corporations survey. - Footnote definition preserved: "Those with deposits or substitutes of deposits representing more than 5% of the total deposits in the current coverage of depository corporations." ### VI. Balance of Payments Statistics — Actions Taken - SDS introduced an advance release calendar for balance of payments statistics indicating no later than dates for preliminary, revised, and final quarterly and annual statistics; calendar to be placed on SDS website. - Calendar will indicate simultaneous release to all users with no prior government access; data initially disseminated via press release and SDS website, subsequently in NBG’s Quarterly Bulletin and SDS Statistical Yearbook. - SDS now disseminates annual balance of payments data on its website. - Balance of Payments Division allocated two new computers. ### VI. Balance of Payments Statistics — Actions Planned - Prerequisites — legal/institutional environment: - SDS will propose balance of payments statistical surveys for inclusion in State Program of Statistics under Law on Statistics mandate. - Subject to Government decision on increased penalties for nonresponse, SDS will apply penalties selectively and publicize them. - Subject to State Department of Customs being made responsible for compiling trade statistics consistent with international standards for coverage and valuation of goods crossing the customs frontier, SDS will work with Customs to reconcile bilateral discrepancies with major partner countries. - Resources: - SDS will propose additional staff and financial resources for balance of payments compilation within the medium-term budget. - Integrity — transparency: - SDS will state on its website that strict and unambiguous embargo arrangements apply to new or revised balance of payments data to ensure simultaneous release; users requested to attribute data to SDS. - SDS will indicate on website that balance of payments data are compiled largely from administrative data sources and that some sources (e.g., customs data for merchandise trade) are seriously deficient for balance of payments purposes; efforts to improve these sources and develop new ones will be reflected in revised estimates in future. - Accuracy and Reliability — source data: - Introduce a program of surveys: a general balance of payments survey covering all large enterprises (major nonmerchandize transactions), a specialized inward foreign direct investment survey (priority), and specialized surveys for services receipts (transport, communication, construction, insurance). - Accuracy and Reliability — statistical techniques: - Develop new techniques to estimate shuttle trade, income and expenditure of residents working abroad, and workers’ remittances. - Serviceability — relevance and consistency: - Establish an advisory group (including agencies consulted in ROSC User Survey) to assess relevance of balance of payments statistics and review how well user needs are met. - Develop a sources and methods guide to inform compilers about compilation practices. - Revision policy and practice: - Include a revision policy statement on SDS website indicating intention to periodically undertake studies documenting reasons for direction and size of revisions; preliminary, revised, and final data will be noted. - First study of revisions to balance of payments statistics over recent years is currently under way. - Accessibility — data and metadata: - Post summary and detailed quarterly and annual balance of payments data on SDS website in a format addressing wide user needs and including a detailed statement corresponding to recommendations of the fifth edition of the Balance of Payments Manual. - Develop an abbreviated sources and methods document for users as an aide; a fuller compilers’ version is being prepared. *Source: Response by the Authorities (IMO: Ministry of Finance, National Bank of Georgia, State Department of Statistics), May 14, 2003.* ### 1. DQAF: Summary Presentation of Results for National Accounts.......................................21 ### 1. DQAF: Summary Presentation of Results for National Accounts ### 0. Prerequisites of Quality — legal, institutional, and resource context - Legal and institutional framework - The State Department of Statistics (SDS) is an independent government institution operating under the Law on Statistics No. 1071 of November 12, 1997. - Regulation for the SDS No. 92 from February 28, 1998, Article 5: SDS is the responsible agency for compiling national accounts statistics. - A State Program for Development of Statistics for 2001–2005 includes compilation and dissemination of national accounts statistics. - Data sharing and coordination - Article 11 of the Law guarantees SDS access to data produced by legal entities, including government ministries and the National Bank of Georgia (NBG). - Written agreements exist between the SDS and the MOF, the NBG and the SDC covering the sharing of macroeconomic data. - No regular contacts are maintained with other statistics producing agencies to address respondent burden. - Confidentiality and respondent protection - Confidentiality guaranteed under Articles 7, 12, and 17 of the Law; the Law does not provide penalties for disclosure. - SDS is planning an amendment to introduce penalties for disclosure. - Anti-monopoly Law permits the State Anti-monopoly Service of MOEIT to request confidential data from SDS; conflict unresolved though no practical breaches reported. - Staff required to sign a declaration on confidentiality; no special aggregation rules exist to guard against unintended disclosure. - Statistical reporting mandate and respondent compliance - Article 11 authorizes SDS to request information from all legal public and private institutions, and individuals; respondents must furnish requested information accurately and timely. - Code of Administrative Offences (1999) provides some penalties for nonreporting but these are considered ineffective and rarely applied for nonresponse to statistical surveys. - SDS does not systematically review response burden; contacts with respondents are very limited. - Resources: staff, training, computing, and budget - National Accounts Division staff: nine. - High staff turnover: during the last two years seven people have left the division. - Staff salaries: about US$20–30 per month; noted as 15 times less than salaries of economists in the NBG. - Qualifications: generally adequate; three economists trained in JVI course in national accounts; one in a Eurostat course; no other formal national accounts training program. - Computing resources: generally adequate, networked computers, hardware and software up to date. - Budgetary constraints: SDS received only about 40–50 percent of amounts budgeted in the last three years. - Financing of key basic economic statistics surveys is dependent entirely on donor aid (examples: household budget survey and labor force survey). - Efficiency and use of resources - National accounts staff are often engaged in other statistical activities, including donor financed ad hoc surveys. - Quality awareness and oversight - Organizational units: Statistical Research Institute (methodology/software) and Institute for Informatics and Telecommunications (information/analytical systems, classification harmonization). - Scientific and Advisory Council established in 1997 including SDS, MOEIT, NBG, academic specialists. - No separate plan for improving quality as such. - No body providing guidance on quality of national accounts statistics or strategies for improvement. - Two user surveys conducted in the last two years focus on general dissemination issues and not on monitoring data quality. - No systematic reviews of problems at various stages of data collection, processing, dissemination due to resource constraints. - Tradeoffs between timeliness and accuracy are not systematically considered; staff spend considerable time meeting tight timeliness for flash quarterly and annual estimates rather than prioritizing preliminary and final estimates. ### 1. Integrity — professionalism, transparency, ethical standards - Professionalism - Law on Statistics emphasizes international standards, definitions, authenticity, objectivity, and independence. - Regulation addresses SDS role as independent data producing agency. - Professional competency influences promotion; methodologies disseminated; participation in international statistical organizations encouraged. - No ministerial interference reported in choice of statistical sources and methods. - NAD selects from a wide range of survey and administrative data. - National accounts staff comment on erroneous interpretation or misuse of statistics; preliminary releases are supplemented with explanatory material for press. - Transparency - Terms and conditions governing collection are published in the Law on Statistics and set out on collection forms. - Publications identify information about SDS but not about its products; SDS publications do not require attribution when statistics are reproduced. - No public information on who has access to data prior to release; MOEIT, MOF, IMF and some other international organizations have access to data as soon as they are available. - Advance public notice is given of major changes in methodology, source data, and statistical techniques in hard copy publications. - Ethical standards - Ethical rules laid down in the Code of Conduct for Civil Servants; new staff are made aware on joining. ### 2. Methodological Soundness — concepts, scope, and coverage - Concepts and definitions - Georgian national accounts are based on the concepts and definitions recommended by the 1993 System of National Accounts (1993 SNA). - Scope and coverage - Minimum requirements established by the Intersecretariat Working Group on National Accounts are not met. - SDS compiles and disseminates: - Value added and gross domestic production or GDP (annual and quarterly) at current and constant (previous year) prices by activity; - Annual expenditure of GDP at current prices; and - Generation of Income Account (annually). - Additional compilations: - Quarterly GDP estimates by the production approach at current and constant prices. - Annual supply and use tables are compiled but not disseminated. - Territory and residence criteria: - National accounts cover the economic territory of the country; residence defined in accordance with the 1993 SNA and BPM5. - Abkhazia and South Ossetia are included in territory but there are significant problems receiving data from these regions. - Workers who work part of the year in neighboring countries (mainly Russia) are included in the national accounts. - Production and assets boundaries: - Production boundary in accordance with 1993 SNA; scope of output includes own-account production for own-final consumption, own-account fixed-capital formation, research and development on own account, mineral exploration, production of computer software, entertainment, literary and artistic originals. - Illegal activities are not included. - Assets boundary generally in accordance with 1993 SNA; agricultural work-in-progress included among tangible assets. - Defense-related assets usable for civilian purposes, valuables and historical monuments are not currently in scope for estimating the assets boundary. - Intangible assets included: mineral exploration; systems and standard applications computer software and databases; entertainment, literary, or artistic originals; patented entities; leases and other transferable contracts. *Source: Detailed assessment using the Data Quality Assessment Framework (DQAF) — National Accounts section.* ### 2.3 Classification/Sectorization ### 2.3 Classification/Sectorization ### Classification and sectorization framework - The 1993 SNA is followed in classifying institutional units, transactions, and other flows. - GDP estimates by production approach are classified according to the General Industrial Classification of Economic Activities within the European Communities (NACE Rev. 1). - The Classification of Products by Activities (CPA) is used for classifying products. - Government expenditure is classified according to the Classification of the Functions of Government (COFOG). - Individual consumption expenditure is classified according to the Classification of Individual Consumption according to Purpose (COICOP). - Foreign trade is classified according to the Harmonized System. ### 2.4 Basis for Recording ### Valuation and recording practices - Valuation rules are in accordance with the 1993 SNA. - Market output is valued at basic prices. - Output for own use is valued at equivalent market prices. - Purchases are valued at purchaser’s values, including trade and transport margins, net taxes on products, and nondeductible value added tax (VAT). If the VAT is deductible, it is not included. - Total exports are recorded on an f.o.b. basis. - Total imports are recorded on a c.i.f. basis and valued on an f.o.b. basis using an estimate for insurance and freight by commodities. - Authorities report that transfer prices have not been detected. ### Recording basis and grossing/netting - Enterprises are asked to provide data on an accrual basis in accordance with the 1993 SNA. - Basic data for central government are on a cash basis. - Expenditures of government for education are adjusted to an “accrual” basis by using the planned budget values for education instead of the cash data. - Transactions between establishments within the same enterprise are recorded on a gross basis. ### 3 Accuracy and Reliability ### 3.1 Source Data — Coverage and main data sources - The SDS maintains a statistics register for nonfinancial enterprises and the Non Profit Institutions Serving Households (NPISH) sector covering about 26,000 enterprises; the register was established from administrative records and the annual survey of revenues and expenditures. - The register contains legal status, type of unit, activity, ownership, economic data, including number of employees, but coverage is not exhaustive because no economic censuses have been conducted for decades. - Main sources for annual national accounts: - Nonfinancial sector and NPISHs: annual survey of nonfinancial enterprises and NPISHs. - General government: data from the MOF. - Financial sector: data on depository corporations from the NBG. - Household sector: household budget survey and ad hoc surveys for unincorporated household activities. - The annual survey of revenues and expenditures covers all activities; coverage in terms of value added: - Industry: coverage considered good. - Transport and communications: coverage satisfactory. - Retail trade, construction, catering, and services: coverage considered poor. - All enterprises are surveyed through complete enumeration; sample surveys applied only for retail trade. - Large enterprises are surveyed by a detailed questionnaire; small and medium enterprises by a simplified questionnaire. - Data collected are sufficiently detailed to derive output, intermediate consumption, fixed-capital formation, and changes in inventories. ### Main deficiencies and nonobserved economy - Significant number of statistical units, mostly unincorporated enterprises, are not surveyed systematically; their share in GDP is about 37 percent. - In agriculture, about 700 thousand household farms account for about 20 percent of GDP and are not surveyed through a systematic annual survey. - Similar undercoverage exists in retail trade, services, and construction. ### Household Budget Survey (HBS) - HBS is a quarterly survey of about 3,300 households. - Sample is based on the population census in 1989 and uses stratified random sampling, stratified for urban and rural areas. - Sample is considered representative at country and regional levels and across socio-economic groups. - Response rate is about 85 percent. - Grossing-up factors are derived based on sample design. - Expenditure items are based on COICOP groups and cover consumption goods and services, durable goods, production for own-consumption, own-account fixed capital formation, and purchases of valuables. - Data on employment in the informal sector are collected for national accounts. ### Government, financial, and external data - Comprehensive government finance statistics available monthly, quarterly, and annually for (i) central government, (ii) local government, and (iii) extra-budgetary funds, compiled by the MOF; local government data are not considered exhaustive. - MOF provides data on all defense related expenditures. - Data on capital stock are not considered comprehensive. - Commercial bank data are received from the NBG and are sufficiently detailed. - Data for external transactions in goods and services come from the SDS Balance of Payments Division; quality is poor due to source data from the State Department of Customs. ### Censuses and periodicity - No regular establishment censuses. Last agricultural census was conducted in 1985. - A census of establishments in Tbilisi was conducted in 2001 with TACIS support. - Population census carried out in January 2002. - SDS collects monthly cumulative output and employment data for nonfinancial enterprises surveyed annually; SDS staff consider these data not quite reliable due to reporting timing (seventh day after month end). - Ad hoc studies conducted mostly at request of and financed by international organizations. ### 3.1.2 Sources approximate national accounts definitions - Abkasia and South Ossetia are included in economic territory, but almost no source data are collected for these regions. - SDS national accounts staff design annual surveys to approximate required definitions, scope, valuation, and classifications. - Fiscal year is calendar year, limiting timing inconsistencies. - Compilers adjust for conceptual differences, but rely heavily on questionnaire instructions and assume enterprise-reported values follow 1993 SNA concepts. ### 3.1.3 Timeliness of source data - Most source data timely except the annual survey of enterprises, which is delayed by three−four months. - Price statistics, balance of payments, monetary and financial statistics, and government finance statistics are available on a timely basis. ### 3.2 Statistical Techniques — Compilation methods and practices - Production approach: output compiled by economic activities at the four-digit NACE level. - Data on intermediate consumption collected only through the annual survey; flash and preliminary GDP estimates use previous year’s ratio of intermediate consumption to output. - Output of owner-occupied dwellings valued as estimated rentals; owner-occupiers asked to self-assess. - Work-in-progress includes growing crops, livestock reared for food, large construction projects, and output of large equipment. - Inventories: respondents requested to value opening and closing stocks at average prices of the year; authorities consider respondents follow instructions and holding gains/losses do not enter value added valuation. - Consumption of fixed capital estimates based on book-keeping values. - Government expenditure, revenue, taxes, and subsidies on products are valued on a cash basis. ### Deflation and volume measurement practices - Double indicator method is not used for compilation of value added of any activity. - Single deflation of value added is applied: - Industry using PPI. - Construction using CPI sub-group for building materials. - Transport and communications using CPI sub-groups. - Services using total CPI. - Agriculture: extrapolation of value added by volume indices. - Volume measures of taxes and subsidies on products estimated by deflation of current values using total CPI. - Output volume of trade margins estimated by deflating current values using CPI sub-groups. - Volume changes measured using annual chain indices for annual data and previous year base indices for quarterly data. - GDP estimates in current prices by expenditure approach are derived independently: - Household final consumption expenditure (one digit COICOP). - Government final consumption expenditure (one digit COFOG). - Gross fixed capital formation (by activities and types of assets). - Changes in inventories (by activities). ### Quarterly compilation techniques - Quarterly estimates compiled from cumulative data. - Quarterly estimates derived from seasonally unadjusted source data. - No seasonal adjustment is performed. ### 3.2.2 Other statistical procedures - Estimated share of the nonobserved economy in the Georgian 2001 GDP is about 33 percent. - SDS uses supply and use tables, HBS data, other SDS surveys, and administrative data to include nonobserved economy components. - Estimates made for underreporting (based on comparing Labor Force Survey and establishment survey employment) and undercoverage (case-by-case studies). - Agriculture output and intermediate consumption estimated using commodity flows, area planted, livestock numbers, and productivity. ### 3.3 Assessment and Validation of Source Data - Local statistical offices collect source data; no validation at local offices. Hardcopy questionnaires sent to headquarters where first validation occurs. - If data considered inaccurate, questionnaires returned to local offices for corrections; lack of resources limits local office follow-up and auditing. - Headquarter staff validate using available software and hardware; high-value transactions confirmed with respondents. - Sampling error information available only for HBS; nonsampling error information available for other surveys. - Source data analyzed for underreporting and temporal consistency; adjustments made for nonsampling errors. - Accuracy of budgetary data, balance of payments data, and price data is routinely assessed. ### 3.4 Assessment and Validation of Intermediate Data and Statistical Outputs - Intermediate data on major activities (mainly industry) assessed against related indicators such as volume and price trends. - Discrepancies of major intermediate data are assessed; ratios checked include: - intermediate consumption/gross output, - gross value added/gross output, - consumption of fixed capital/gross output. - Value added structure by industries checked for temporal consistency; adjustments made to remove discrepancies. - Statistical discrepancy for 1999 GDP by production and expenditure approaches is about 10 percent and is shown explicitly. - Supply and use tables (with a two year lag) are used to improve consistency; after SUT application, discrepancy is about 3 percent. - No studies about bias in the GDP estimates are performed. ### 3.5 Revision Studies - Direction and magnitude of revisions between preliminary and final data are investigated. - There is no documentation of sources of errors, omissions, and fluctuations. - Results of investigations and adjustments are taken into account in compiling subsequent period data. ### 4 Serviceability ### 4.1 Relevance - SDS assesses whether economic statistics respond to government needs through consultations mainly with government agencies and some academic institutions. - SDS conducted two user surveys in the last two years. - No advisory group or established process to systematically assess relevance of existing national accounts data. ### 4.2 Timeliness and periodicity - Annual national accounts estimates disseminated within the GDDS recommended timeliness limit of six–nine months after the reference year. - Flash estimate of GDP in current and constant prices available one month after the end of the year. - Preliminary annual GDP figures at current and constant prices published after four months. - Final annual national accounts estimates in current and constant prices available 13 months after the end of the year. - Quarterly GDP flash estimates (cumulative) published one month after the reference period; preliminary quarterly estimates published four months after the reference period. - Dissemination periodicity for annual GDP statistics meets GDDS recommendations. ### 4.3 Consistency - Internal consistency issues: - Statistical discrepancy between GDP by activity and expenditure components is not stable over time. - Estimates at constant prices are available only through the production approach. - In supply and use tables, supply does not match total uses and is balanced through changes in inventories. - Quarterly and annual GDP estimates are conceptually consistent; fourth quarter cumulative estimates equal annual estimates for flash and preliminary annual GDP. - Time series availability: - Annual national accounts estimates available since 1990; quarterly estimates since 1996. - Two breaks in annual accounts: - Currency/classification: expressed in rubles 1990–1992, Georgian coupons 1993–1994, and in Lari from 1995. - Classification change: old classification until 1996, NACE since 1996. - Consistent annual time series available since 1996; historical annual series were not reconstructed for methodological changes. - Quarterly estimates are consistent over time. - Detailed methodological notes explain breaks in time series; SDS explains unusual changes in analytical text. ### Reconciliation with other datasets - National accounts statistics reconcilable with general government data produced by the MOF. - Consistency checks between national accounts and the balance of payments are not possible because SDS compiles only Production and Generation of Income Accounts. - Data on imports and exports of goods and services for GDP by expenditure approach are consistent with balance of payments current account data. *Source: _cr03140 - 2.3 Classification/Sectorization* ### 4.4 Revision Policy and Practice ### _cr03140 - 4.4 Revision Policy and Practice ### 4.4 Revision policy and practice — main findings - Revisions follow an established schedule for national accounts annual estimates by the production approach: - Flash estimate—one month after the reference period - Preliminary estimate—four months after the reference period - Final estimate—13 months after the reference period - The authorities do not follow a regular schedule for revisions of GDP by the expenditure approach. - Quarterly estimates are revised following the same pattern as the annual production-approach schedule. - The SDS does not provide extensive documentation on its major revisions. - New statistical techniques are sometimes introduced into the accounts without revising earlier data, creating breaks in series. - Flash, preliminary, and final estimates are clearly identified; users are made aware that data are subject to revision. - Revised data are disseminated with the same level of detail as previously published data. - Major revisions due to changes in methodology are measured and explained in the press release when disseminated, but explanations do not include analysis of preliminary versus revised data for the major aggregates that were revised. ### 5 Accessibility — Data accessibility (5.1) - Presentation and dissemination practices: - National accounts estimates are disseminated in the SDS Statistical Yearbook of Georgia (SDS Yearbook), the Monthly Statistical Bulletin and through press releases. - Annual current price estimates and growth rates are published by economic activity at the one digit level of NACE and by major expenditure category, with a breakdown by household consumption, government consumption, and capital formation. - Constant price estimates and the deflators of GDP are not published. - Annual estimates are not disseminated at a detailed level; there are no shares, charts, and tables to facilitate analysis. - Final quarterly estimates are not published. - Dissemination media and formats: - Annual and quarterly flash and preliminary estimates are disseminated first by press release and later through the Monthly Bulletin. - Final annual estimates are released only through the SDS Yearbook. - The newly developed web page of the SDS does not contain any national accounts data. - Longer time series are not accessible. - Release scheduling and simultaneity: - Annual flash and preliminary and quarterly national accounts estimates are released according to a pre-announced schedule published each quarter and available to the public. - There may be delays due to the printing process of hardcopy publications. - Final annual estimates are not released according to a pre-announced schedule. - Perception exists that as soon as data are compiled they are considered released; in practice data are released first to the MOEIT, MOF, the NBG, and some international organizations and after two–three days to other users. - The press is not briefed in advance. - Access to unpublished aggregates: - Unpublished but nonconfidential detailed data are made available upon request, but the availability of these data is not publicized. ### 5.2 Metadata accessibility - Documentation and scope: - There is not a detailed description of concepts, definitions, classifications, data sources, and statistical techniques used for national accounts compilation. - SDS has published a short description of the national accounts metadata and a hardcopy description on the measurement of the nonobserved economy. - No metadata is available on the newly created web page of the SDS. - Levels of detail: - Metadata are not made available at different levels of detail. ### 5.3 Assistance to users - Contact information and enquiries: - All statistical releases and publications indicate the mailing address, telephone and fax numbers, and e-mail address of the SDS Publication Division and the specific divisions (including National Accounts Divisions) responsible for compilation. - National accounts staff handle all enquiries and provide assistance to users. - Catalogues and charges: - A list of publications is given in the SDS catalog of publications and on the web page with information on how they may be obtained. - The cost of the publications is not stated. ### Data Quality Assessment Framework — selected assessments (Table 1, summary) - Prerequisites of quality: - 0.1 Legal and institutional environment — assessment: NO; comment: Penalties for noncompliance are ineffective. - 0.2 Resources — assessment: NO; comment: Inadequate staff and financial resources, high staff turnover, unattractive salaries. - 0.3 Quality Awareness — assessment: NO; comment: No special body to provide guidance on quality of national accounts data. - Integrity: - 1.1 Professionalism — assessment: O - 1.2 Transparency — assessment: NO; comment: Internal government access to NA data prior to release is not publicized. - 1.3 Ethical Standards — assessment: O - Methodological soundness: - 2.1 Concepts and definitions — assessment: NA - 2.2 Scope — assessment: O; comment: Volume measures of expenditure GDP are not compiled, only production and generation of income accounts are compiled. - 2.3 Classification/Sectorization — assessment: NA - 2.4 Basis for recording — assessment: O; comment: General government transactions are recorded on cash basis. - Accuracy and reliability: - 3.1 Source data — assessment: NO; comment: Coverage of register is not comprehensive, poor coverage of trade and services, no census surveys have been conducted. A Labor Force Survey and a Census of Agriculture are planned for 2003. - 3.2 Statistical techniques — assessment: NO; comment: Serious deficiencies in techniques for deriving constant price estimates. - 3.3 Assessment and validation of source data — assessment: NO; comment: No validation of source data by the local offices. - 3.4 Assessment and validation of intermediate data and statistical outputs — assessment: NO; comment: Supply and Use Tables are not sufficiently detailed to allocate the discrepancy of GDP between the two approaches. - 3.5 Revision studies — assessment: NO - Serviceability: - 4.1 Relevance — assessment: NO; comment: No systematic review process to assess whether NA meet users’ needs. - 4.2 Timeliness and periodicity — assessment: NA - 4.3 Consistency — assessment: NO; comment: Historical series are not reconstructed. - 4.4 Revision policy and practice — assessment: NO; comment: Revision schedule is not publicized. - Accessibility: - 5.1 Data accessibility — assessment: NO; comment: NA constant price estimates and deflators are not published, dissemination formats are not adequate, there is no set embargo date, data are not released simultaneously. SDS plans to disseminate NA data through its web page. - 5.2 Metadata accessibility — assessment: NO; comment: Very limited documentation on sources and methods. - 5.3 Assistance to users — assessment: NO; comment: The cost of the publications is not stated. ### Consumer Price Index — relevant institutional and resource details - Legal and institutional framework: - SDS operates under the Law on Statistics No. 1071 of November 12, 1997, with an amendment in June 1999. - A decree by the Council of Ministers on the State Service for Registering Prices and Tariffs No. 125, March 10, 1995, established the prices division as a separate unit of SDS. - The Regulation for the SDS No. 92 from February 28, 1998 (Article 21) explicitly mentions the collection and compilation of consumer and producer price statistics. - A State Program for Development of Statistics for 2001–2005 includes the compilation and dissemination of price statistics. - Data sharing and coordination: - Article 11 of the Law guarantees SDS access to data produced by legal entities, including government ministries and the National Bank of Georgia (NBG). - Written agreements exist between the SDS and the MOF, the NBG and the Customs Committee covering sharing of macroeconomic data. - Confidentiality: - Confidentiality is guaranteed under Articles 7, 12, and 17 of the Law; the Law does not provide penalties for disclosure. - SDS plans a new amendment to the Law with a provision for penalties for disclosure. - The Anti-monopoly Law allows the State Anti-monopoly Service to request confidential data from the SDS; the conflict between the two Laws has not been resolved. - Reporting mandates: - Article 11 authorizes SDS to request information from legal public and private institutions and individuals; the June 1999 amendment provides some penalties for nonreporting, but these are not considered effective in practice. - SDS does not systematically review respondent burden; contacts with respondents are limited. - CPI/PPI staffing and resources: - CPI and PPI are jointly compiled by nine staff in the central office in Tbilisi. - In the Tbilisi regional office there are 20 price collectors who spend about 80 percent of their time on prices work. - In the other four regional offices, there is a total of 18 price collecting staff. - Remuneration has fallen to about 7 percent of the 1988 levels in real terms. - Nine central office prices staff have tertiary qualifications in economics; three of them have over 15 years of statistical experience each. - Ninety percent of the regional office price collectors have tertiary qualifications. - Training, computing, and communications: - Internal training covers sampling, methodology, data control, price index theory, maintenance of product quality, treatment of temporarily missing items. - Funding restrictions limit external training; director attended IMF seminars in Vienna (2002) and Kiev (1998) and is recommended to attend the next ILO meeting in 2003. - Computing resources are described as very inadequate: nine Central Office staff share three old personal computers; at least three additional machines are required. Some regional offices lack machines entirely. - CPI software is MS Excel files provided by the IMF upgraded locally; new software is being developed using MS Access. - PPI software needs updating; communications systems are very poor and unreliable with no electronic data transfer of unit records from regional to central office. - Financial resources are very scarce; even stationery and office consumables are in shortage. ### Quality management and improvements — practices and recommendations noted - Management and institutional arrangements: - SDS includes a Statistical Research Institute and an Institute for Informatics and Telecommunications. - A Scientific and Advisory Council was established in 1997 with members from SDS, MOEIT, the University, NBG, and the Institute of Economics. - The mission statement emphasizes improvement and harmonization of statistical activities and the maintenance of independence, accuracy, relevance, integrity, timeliness, and professional standards. - Quality assurance and processes: - Before CPI and PPI release, rigorous output editing and credibility checks are undertaken including cross-checking between CPI and PPI; SDS senior management clears data before each month’s release. - No separate unit provides guidance on quality of price statistics; no systematic reviews to identify problems at collection, processing, and dissemination stages. - Regular meetings with users, particularly the NBG, provide external feedback. - Efficiency measures and opportunities: - SDS management scrutinizes the work program annually and monitors resource usage. - Opportunities recommended: greater use of computing technologies, rationalization of collection forms, upgrading software, exploring efficient data transmission mechanisms, and evaluation of methodologies and compilation techniques with external expert assistance. - Attendance at international meetings and training is used to share experiences and develop best practice. *Source: _cr03140 - 4.4 Revision Policy and Practice (Compiling Agency: State Department of Statistics).* ### 1. Integrity ### _cr03140 - 1. Integrity ### Integrity and Professionalism - The Statistics Law defines principles and responsibilities for collection, processing, and dissemination of statistics in Georgia, emphasizing international standards, authenticity, objectivity, and independence. - Respondents are obliged to provide clear, objective, and timely data to SDS within assured confidentiality. - SDS has complete independence and autonomy in performing its statistical responsibilities; no scope for external interference in compilation and dissemination. - Price statistics are collected and compiled impartially by professional SDS staff with tertiary qualifications and training in statistical techniques; collection and compilation follow international standards and objective statistical judgment. - Weighting sources for price indices are based on household and industrial surveys; design of price collections is based on efficiency, cost effectiveness, data reliability, and respondent burden. - Media monitoring and press conferences are used to correct erroneous interpretation or misuse of statistics; specialist journalist meetings are held regularly. - Example corrective action: public comment made to rectify misinterpretation related to the introduction of cigarette excise tax. ### Transparency - Terms and conditions governing collection are published in the Law of Georgia on Statistics, including obligation to report, penalties for noncompliance, and confidentiality; these are set out on collection forms and discussed with users and journalists as appropriate. - Statistical publications do not identify where more information about the data providing agency and its products can be found. - No explicit public release embargos; in practice, special delivery arrangements for selected government agencies result in those agencies receiving data prior to other users; these arrangements are not publicly acknowledged. - SDS publications are clearly identified by department name and logo; when SDS data are released in another agency’s publication (e.g., NBG), the data are clearly sourced. - SDS publications do not request attribution when its statistics are used or reproduced. - Advance notice is given to users prior to introduction of changes to methodology, sources, or techniques (applies to periodic reweighting of price indices). ### Ethical Standards - SDS operates under a solid legal framework being further strengthened for confidentiality of data, penalties for provider noncompliance, and penalties for offenses by SDS staff. - Civil service law contains statements on ethical standards; SDS guidelines cover handling of confidential data. - Induction training of new SDS staff emphasizes ethical standards. ### Methodological Soundness — Concepts and Scope - CPI developed with IMF technical assistance in 1992 and 1997; compiled according to ILO guidelines and Eurostat recommendations; basic concepts and definitions for household consumption consistent with the 1993 SNA. - CPI covers price change of goods and services consumed by resident households; expenditure weights based on Household Budget Survey (HBS) data for 1996–97. - Prices included are fully specified in product and transaction characteristics. - Weighting pattern to be updated at the end of 2002 using results of the 2001–2002 HBS. - Scope: all resident households living in the five major urban areas; prices collected for five cities with separate weighting for Tbilisi and a common weighting pattern for the other four cities; population-based weights aggregate the five city indices. - Own-account production of market goods for own final consumption is included in CPI weights; owner-occupied dwelling services are not currently included. - Classification: CPI structure based on ILO guidelines; commodities classified according to CPA; COICOP to be introduced with next reweight at end of 2002. - Basis for recording: weights and prices based on actual market transaction prices inclusive of trade and transport margins and taxes; product specifications rigorously defined and product list updated annually. - Recording: prices recorded on an accrual basis; most collected each month between the 10th and the 20th of the month. - Frequency: food, fuel, and gas prices collected three–five times per month; postal, communications, and water prices collected quarterly. ### Methodological Soundness — Classification and Aggregation - CPI built from 296 index items aggregated to eight main publication groups: Food, beverages and tobacco; Clothing and footwear; Rent, water, fuel and power; Household goods; Medical care; Transportation and communication; Recreation, education and culture; Personal care and effects. - Elementary aggregate index numbers calculated from short-term ratios of average prices; modified Laspeyres aggregation formula used. - Weighting base period is 1996–97 with weights adjusted to December 1997 prices (price reference period for the current index). - Weights previously based on 1992, 1994, and 1996–97; link factors derived from December index overlap applied when new weights introduced to maintain continuous historical series. - Temporarily missing prices imputed using current movement in the index at the next level of aggregation; permanently disappeared products replaced; out-of-season prices repeated until new season prices available. - Quality change practice: changed product treated as new item and associated price change spliced out rather than directly estimating value of quality change. - Item list reviewed annually and new items added as appropriate. ### Accuracy and Reliability — Source Data (HBS and Price Surveys) - HBS 1996–97 (with World Bank assistance) used for CPI structure and weighting; HBS conducted on continuous basis each quarter since 1996. - Sample design: two-stage stratified sample using 1989 population census (progressively updated with 2002 census); dwelling is collection unit; 3,351 households surveyed on an eight-quarter rotation basis. - Rotation and monthly coverage: 12 rotation groups; each month 1,117 households are surveyed. - Response rates: about 85 percent in 1997–98; falling to 75–79 percent in 2000–01; increasing towards 85 percent again from 2001–02 after revised collection and questionnaire strategy. - HBS geographic coverage: whole country; market value of goods produced for own consumption covered; rental value of owner-occupied dwellings not measured. - COICOP used from 2002; surveys 1996–2001 based on old Soviet classification. - Nonresponse treatment: - (i) weight for household refusals allocated to other households in same strata; - (ii) if no one living at selected address, weight for that dwelling deleted and actual household weight adjusted. - CPI consumer price survey: prices for most goods and services collected monthly by personal price collection; volatile items (fresh fruit and vegetables, fuel and gas) collected three–five times per month; regulated items (communications, water) collected quarterly. - Outlet sampling: drawn from list of markets, shops, supermarkets developed from Retail Surveys; historically 80 percent of foodstuffs purchased from farmers’ markets; outlet sample selected randomly from list stratified by outlet type, retail industry, and region (e.g., Tbilisi divided into ten administrative regions). - Challenges: informal market outlets make rigorous specification pricing difficult; concerns about adequacy of HBS and price surveys reflecting nonobserved economy activity. - New goods monitored via advertisements and press articles. ### Accuracy and Reliability — Data Timeliness, Compilation, and Validation - Detailed annual data from continuous HBS available two months after year end. - Consumer price quotes normally timely; delays in transferring data to central office can cause missed preannounced release dates due to unreliable communication systems and inadequate financial resources. - Expenditure estimates from HBS produced at sufficiently detailed level for CPI weights for goods; services less well-categorized until COICOP introduction. - Goods for own final consumption covered and valued at market prices at time of production; owner-occupied dwelling services not included. - Imputation and replacement practices for missing or discontinued items described above. - Source data assessment: - HBS response rates closely monitored; when rates declined 1998–2001, collection strategy and form were reviewed. - Standard errors continuously reviewed; successive annual HBS data compared and analyzed. - Price collection uses judgmental sampling; sampling error information not available; nonresponse not an issue due to high respondent cooperation and personal collection. - Long-term analyses of price changes undertaken; relationships within groups studied; outliers investigated and corrected. - Close contact with State commission for regulation on energy and communications for cross checking. - Intermediate data validation: - Price movements compared at consumer and producer level; aggregate CPI and PPI movements nearly identical over the last six months. - Large or inconsistent movements investigated; exchange rate impacts on imported consumer goods evaluated; prices of regulated items verified. - Comparisons between five regional indices undertaken; imputations made consistently; overall CPI for Georgia aggregated from five regional indices with no discrepancies reported. ### Revisions and Historical Continuity - CPI is final when first released; index numbers are not subsequently revised. - One instance of a revision in eight-year CPI history due to an error in timing of recording an increase in railroad tariffs. - When weights or methodology are revised, index series are not backcast and revised; new series are linked onto old series to form continuous historical series. - Prior to weight updates, detailed analysis of consumption shifts undertaken and major changes explained. ### Serviceability and Relevance - Ongoing consultation with users, especially major public policy organizations; press conferences and briefings with journalists used to obtain feedback. - No formal user group nor established systematic review process to assess how well users’ needs are met. - SDS representatives attend international meetings subject to available financial resources. *Source: _cr03140 - 1. Integrity (Consumer Price Index), IMF PDF* ### 4.2 Timeliness and periodicity ### 4.2 Timeliness and periodicity ### Timeliness and periodicity - The timeliness of the price indices meets GDDS dissemination standards. - The CPI is released ten days after the end of the reference month. - The CPI is released monthly and therefore meets GDDS dissemination standards. - There are additional lags before publications are available to users. ### Continuity and historical coverage - Continuous monthly, quarterly, and annual index series are maintained through processes of linking at the time new weights are introduced. - CPI series go back to January 1994. ### Internal and external consistency - The CPI is internally consistent; the all-items CPI is invariant to the geographic or commodity classifications. - The CPI and PPI are constructed on a consistent basis, and are compatible with national accounts statistics to the extent that they adhere to 1993 SNA concepts and methods. ### Revision policy and practice - The CPI is not revised and the monthly index numbers are final when first released (note the one exception mentioned under 3.5.1). - This “no revision” policy is well known to the public. - Weights are updated on a regular basis, and product and outlet samples are modified over time; index series are not backcast and revised to reflect new weights or samples. New indices are linked onto the previous series to form a continuous historical time series. - Preliminary data identification: Not applicable, as index numbers are final when first released. - Studies and analyses of revisions: Not applicable. ### Accessibility — Data dissemination and formats - CPI publication contains clear tables and charts and a reasonable amount of detail in time series form; however, no analysis of current period changes is presented. - As soon as data are finalized, they are considered released and are available to anyone who requests them. - Data release sequence: first by fax and e-mail to selected organizations, subsequently to the media, later in a monthly SDS General Report, and then in a more comprehensive specialist quarterly CPI publication. - The data are not currently released electronically. - There is a pre-announced schedule of publication release dates (but not times); the target release date for the CPI is the tenth of the month following the reference month (subject to when weekends fall). In practice, the dates are not always met. - In principle all users have access simultaneously; in practice selected government users and international organizations receive data before other users. - There is no strict and well-publicized embargo time (for example, 10:00am), contributing to inequitable treatment of users. - Nonconfidential, unpublished index numbers are made available on request; the availability of this service and the terms and conditions are not publicized. ### Accessibility — Metadata and user assistance - Public metadata are inadequate: the only metadata made available to the public is a very brief description of the CPI contained in the SDS Yearbook. - There is no concepts, sources, and methods document or other descriptive material about the CPI. - A contact person and telephone number is provided in the fax sent to public policy users and on the cover sheet accompanying the CPI publication; a prompt and knowledgeable service is provided to enquirers. - There is a catalogue of publications but it is not comprehensive and does not contain charges. ### Summary operational and resource facts relevant to timeliness and serviceability - The compilation and dissemination of price statistics are stated in the State Program for Development of Statistics for 2001–2005. - The State Department of Statistics (SDS) operates under the Law on Statistics No. 1071 of November 12, 1997, supplemented by an amendment in June 1999. - A decree by the Council of Ministers of Georgia established the prices division as a separate unit of SDS: State Service for Registering Prices and Tariffs No. 125, March 10, 1995. - The collection and compilation of consumer and producer price statistics are mentioned in the Regulation for the SDS No. 92 from February 28, 1998. - Staffing and computing resources for CPI and PPI: - Nine staff in the central office in Tbilisi compile CPI and PPI. - In the Tbilisi regional office there are 20 price collectors who spend about 80 percent of their time on prices work. - In the other four regional offices there is a total of 18 price collecting staff. - The nine central office prices staff share three old personal computers; at least three additional machines are required. - The PPI and CPI software: CPI software based on MS Excel files provided by the IMF and upgraded locally; new software being developed using MS Access. PPI software needs updating. - Remuneration is extremely low; SDS salaries have fallen to about 7 percent of the 1988 levels in real terms. - 90 percent of the regional office price collectors have tertiary qualifications. - Communications systems are very poor and unreliable; no electronic data transfer of unit records from regional offices to central office — data are physically delivered by courier or SDS staff. As a result, sometimes the pre-announced CPI and PPI publication release dates are not met. - Financial resources are very scarce; there are shortages in travel, course attendance, stationery, and other office consumables. *IMF staff assessment as presented in the source document.* ### 1. Integrity ### 1. Integrity ### 1.1 Professionalism - The Statistics Law defines principles and responsibilities for collection, processing, and dissemination of statistics, emphasizing international standards, authenticity, objectivity, and independence; respondents are obliged to provide data to SDS with assured confidentiality. - SDS has complete independence and autonomy in performance of statistical responsibilities; no scope for external interference in compilation and dissemination. - Price statistics are collected and compiled impartially by professional SDS staff with tertiary qualifications and training in statistical techniques; collection and compilation conform to international standards and apply objective statistical judgment. - Weighting sources for the price indices are based on household and industrial surveys; design of price collections is based on efficiency, cost effectiveness, data reliability, and respondent burden. - Media monitoring and public comment are used to correct major misinterpretation of price indices (example: impact of the introduction of cigarette excise tax). - Press conferences and regular meetings with specialist journalists accompany major releases and methodological discussions. ### 1.2 Transparency - The Law of Georgia on Statistics publishes the terms and conditions governing collection of statistics, including obligation to report, penalties for noncompliance, and confidentiality; these conditions are also set out on collection forms and discussed with users and journalists. - Statistical publications do not identify where more information about the data producing agency and its products can be found. - In practice, lack of specific public release times and special delivery arrangements for selected government agencies result in those agencies receiving data prior to other users; these arrangements have not been publicly acknowledged. - SDS publications are clearly identified by department name and logo; when SDS data appear in another agency’s publication (e.g., price indices in an NBG publication) the data are clearly sourced. - SDS does not state in publications that attribution should be made to it when its statistics are used or reproduced. - Advance notice is given prior to introduction of changes to methodology, sources, or techniques (applies to periodic reweighting of price indices). ### 1.3 Ethical Standards - SDS operates under a solid legal framework being strengthened for confidentiality, penalties for noncompliance by providers, and penalties for offenses by SDS staff. - Civil service law contains statements on ethical standards; SDS guidelines cover handling of confidential data. - Induction training of new SDS staff emphasizes ethical standards. ### 2. Methodological Soundness #### 2.1 Concepts and Definitions - The monthly PPI is compiled using 1993 SNA concepts for recording and valuation of producer prices and value weights. - The index is a gross sector output index with weights based on industry gross output value data from the Enterprise Survey for 2001, valued at basic prices. - Prices relate to market transaction prices valued at basic prices, excluding trade and transport margins and taxes on products. - Prices are transaction prices inclusive of discounts and any rebates; product specifications cover product and transaction characteristics. #### 2.2 Scope - Scope is confined to output of Mining, Manufacturing and Electricity, gas and water sectors (NACE Sections C, D, and E). - All resident market enterprises for the defined sectors are in scope; production for own account use is not significant. #### 2.3 Classification/Sectorization - Classification conforms to 1993 SNA; NACE used for industries and CPA used to classify commodities. - Product descriptions allow unique classification to a CPA category and enable collection of prices of the same products month to month. #### 2.4 Basis for Recording - Valuation of PPI weights and prices is at basic prices consistent with 1993 SNA; collected prices are actual market transaction prices including discounts, excluding trade and transport margins and taxes on products. - Product specifications include all price-determining characteristics to support constant-quality pricing. - Prices for goods are recorded on an accrual basis for the period in which they are sold; prices are collected monthly and included in the index for that month. - PPI is a gross sector index and not categorized by Stage of Production (3.4.3 not applicable). ### 3. Accuracy and Reliability #### 3.1 Source Data - Business Register is inadequate as a frame for PPI price collection because it does not differentiate between operating and nonoperating enterprises and does not accurately reflect types of industrial activity. - PPI uses a list of operating enterprises obtained each year from the quarterly Enterprise Survey (Industry Division of SDS) covering enterprises that operated continuously during the previous four quarters, categorized by industry, product, and value of output. - PPI output weights are based on annual Enterprise Survey results covering Manufacturing, Mining and Electricity, gas and water. - Enterprise Survey coverage: covers all large enterprises in the defined sectors, all enterprises that hire employees, and a random sample of enterprises that do not hire employees (10 percent of the enterprises in an industry with a large number of enterprises). - The Enterprise Survey covers 2,500 of the 3,000 operating enterprises, so coverage of output is extremely high. - Calendar year results from the Enterprise Survey are available by mid-January. - The monthly producer price survey uses a shuttle form covering 12 months; at the beginning of each year regional offices receive forms for the new sample and fully specified products. - PPI sample includes all major units and selection of others using cut-off sampling; a minimum of 60 percent coverage is achieved for each industry; total of 260 units included in the PPI collection. - Producer prices for large enterprises and problematic reporters are personally collected by regional staff; other enterprises and rural units are collected by mail. - There are reservations about adequately reflecting the size of the nonobserved economy in weights and prices. - Enterprise Survey data and producer prices are classified by NACE and are consistent with PPI requirements. - Producer prices are due on the fifth of the month following the reference month; data are normally timely though transfers to central office can delay publication meeting the preannounced release date due to unreliable communication systems and inadequate financial resources. #### 3.2 Statistical Technique - Gross output value estimates from the Enterprise Survey are produced at sufficient detail for PPI weights. - PPI structure relates to 83 three-digit NACE Classes, 28 Groups, 16 Subsections and three Sections. - Elementary aggregate index numbers for CPA commodities calculated from short-term ratios of average prices; modified Laspeyres aggregation formula used. - Weighting base period is 2000 with weights adjusted to prices of December 2001 (price reference period for current index). - Weights updated annually; new weights introduced using link factors from a December index overlap to form continuous historical series. - Missing prices imputed using current movement in the index at the next level of aggregation; permanently disappeared products are replaced. - No seasonal products encountered to date; if encountered they will be repeated until new season prices are available. - Quality changes: changed product treated as a new item and associated price change spliced out rather than directly estimating quality change value. - Item list reviewed annually and new items added as appropriate. #### 3.3 Assessment and Validation of Source Data - Main source data are (i) quarterly Enterprise Survey (gross output by industry, commodity, region for annual weighting) and (ii) monthly survey of producer prices. - Enterprise Survey response rates are traditionally very high; estimation of missing quarterly observations by reference to prior quarters; coverage so high that standard errors are not an issue. - In forming new PPI weighting pattern, Enterprise Survey results are analyzed in detail at product level through commodity and industry hierarchy; strict quality control and coding/reporting queries undertaken. - Personal interviews establish monthly price collection; for each commodity four most important specifications in terms of sales value are selected in consultation with the enterprise. - Monthly producer price collection has rigorous procedures to minimize errors; price collectors query price changes in the field and obtain reasons; nonresponse not an issue due to high respondent cooperation and use of personal collection. - Long-term analyses of price changes are undertaken; relationships within groups and subgroups are studied and shifts investigated; individual prices checked against external sources where possible. - Price ratios calculated and products clustered by ranges to ensure consistency; outliers investigated. #### 3.4 Assessment and Validation of Intermediate Data and Statistical Outputs - Price movements for domestically produced consumer products are compared at producer and consumer level; aggregate CPI and PPI movements have been nearly identical over the last six months. - Large movements and inconsistent movements between series are investigated, especially for important commodities such as oil and coal. - Aggregate PPI is a single hierarchical Laspeyres aggregation of commodity and industry indices with no scope for statistical discrepancies. #### 3.5 Revision Studies - The PPI is final when first released and index numbers are not subsequently revised. - When weights or methodology are revised, index series are not backcast and revised; new series are linked onto old series to form continuous historical series. - Prior to updating weights, detailed analysis of shifts in production pattern is undertaken and major changes explained. ### 4. Serviceability #### 4.1 Relevance - Ongoing consultation with users on price indices, especially major public policy organizations; press conferences and briefings with journalists obtain feedback. - No formal user group and no established process to systematically assess how well users’ needs are being met. - SDS representatives attend international meetings within financial resource constraints. #### 4.2 Timeliness and periodicity - Timeliness meets GDDS dissemination standards: PPI released 20 days after the end of the reference month; however, additional lags may occur before publications are available to users. - Periodicity meets GDDS dissemination standards: PPI is released monthly. #### 4.3 Consistency - PPI is internally consistent with a single hierarchical index structure. - Continuous monthly, quarterly, and annual index series are maintained through linking when new weights are introduced; PPI series go back to January 1997. - PPI and CPI are constructed on a consistent basis and are compatible with national accounts statistics to the extent they adhere to 1993 SNA concepts and methods. - Scope of the PPI relates to less than 20 percent of GDP. #### 4.4 Revision Policy and Practice - PPI monthly index numbers are final when first released; this policy is well known to the public. - Weights are updated regularly and product and outlet samples modified over time to remain representative; new indices are linked onto previous series rather than backcasting and revising historical series. - Preliminary data identification and studies/analyses of revisions are not applicable since index numbers are final when first released. *Source: _cr03140 - 1. Integrity* ### 5. Accessibility ### 5. Accessibility ### 5.1 Data Accessibility - Presentation and clarity - The PPI publication contains clear tables and charts. - A reasonable amount of detail is published and it is in time series form. - However, no analysis of current period changes is presented. - Dissemination media and formats - As soon as the data are finalized, they are considered to be released and are available to anyone who requests them. - Data are first released by fax and email to selected organizations and subsequently to the media. - The data are later released in a monthly SDS General Report and then in a more comprehensive specialist quarterly PPI publication. - The data are not released electronically. - Release schedule and simultaneity - There is a pre-announced schedule of publication release dates (but not times), though, in practice, the dates are not always met. - It is reasonably well known amongst users that the target release dates for the PPI is the 20th of the month following the reference month (subject to when weekends fall). - In principle, all users have access simultaneously. However, the release mechanisms result, in practice, in selected government agencies receiving the data before other users. - The lack of a strict and well-publicized, embargo time (e.g., 10:00am) contributes to this inequitable treatment. - Availability of nonpublished aggregates - Nonconfidential, unpublished index numbers are made available on request. - The availability of this service is not publicized, nor the terms and conditions that apply. ### 5.2 Metadata Accessibility - Documentation on concepts, scope, classifications, basis of recording, data sources, and statistical techniques - The only metadata made available to the public is a very brief description of the PPI contained in the SDS Yearbook, which is completely inadequate. - Levels of detail and audience adaptation - There is no concepts, sources, and methods document or other descriptive material about the PPI. ### 5.3 Assistance to Users - Contact points and responsiveness - A contact person and telephone number is provided in the fax sent to public policy users and on the cover sheet accompanying the PPI publications. - A prompt and knowledgeable service is provided to enquirers. - Catalogues and fees - There is a catalogue of publications but it is not comprehensive and does not contain charges. ### Table 3 — Data Quality Assessment Framework: Summary Presentation of Results for Producer Price Index - Key to symbols: NA = Not Applicable; O = Practice Observed; LO = Practice largely Observed; LNO = Practice Largely Not Observed; NO = Practice Not Observed; SDDS = Complies with SDDS Criteria - Selected assessments and comments (as presented) - 0. Prerequisites of quality - 0.1 Legal and institutional environment: NO — Penalties for noncompliance are ineffective. - 0.2 Resources: NO — Computer and financial resources are inadequate. - 0.3 Quality Awareness: NO — There is no separate unit to provide guidance on quality. - 1. Integrity - 1.1 Professionalism: NO - 1.2 Transparency: NO — The public is not made aware of dissemination practices that result in selected government agencies receiving access in advance of other users. Information on the agency and its products is not publicized. - 1.3 Ethical Standards: NO - 2. Methodological soundness - 2.1 Concepts and definitions: NO - 2.2 Scope: NO - 2.3 Classification/Sectorization: LNO - 2.4 Basis for recording: NO - 3. Accuracy and reliability - 3.1 Source data: NO — The business register is not adequate as a frame for the PPI. The nonobserved economy may not be adequately reflected in the weights and price collections. - 3.2 Statistical techniques: NO — Quality change assessment techniques need improving. - 3.3 Assessment and validation of source data: LNO - 3.4 Assessment and validation of intermediate data and statistical outputs: LNO - 3.5 Revision studies: NO - 4. Serviceability - 4.1 Relevance: NO — There is no review process that systematically assesses whether the PPI is meeting users’ needs. - 4.2 Timeliness and periodicity: NO - 4.3 Consistency: LNO - 4.4 Revision policy and practice: NO - 5. Accessibility - 5.1 Data accessibility: LNO — Data are not released electronically. There is no set embargo time. Data are not released simultaneously as government users receive data in advance of other users. - 5.2 Metadata accessibility: NO — There is no adequate meta data documentation. - 5.3 Assistance to users: LNO — The catalogue of publications is incomplete and does not include charges. *Source: _cr03140 - 5. Accessibility (Producer Price Index), State Department of Statistics, as presented in the supplied content.* ### 1.2 Transparency ### 1.2 Transparency ### 1.2.1 Terms and conditions for statistics - Current compilation of fiscal data takes place under the Law on Budget Systems and Powers (No. 249-IIs of May 29, 1999) and annual budget laws, primarily for budget monitoring and management purposes. - Information about this law is freely available to the public. - The terms and conditions under which GFS (as opposed to budget execution data) are collected, compiled, and disseminated by the PRD are not readily available to the public. - It is not publicised that there are no restrictions on public access to GFS. ### 1.2.2 Internal governmental access prior to release - GFS are published in the SDS and NBG dissemination media not later than two months after the end of the reference month. - Prior to these publications: - Fiscal data/GFS flow freely among government agencies, and to individual users outside the government sector about three weeks after the end of the reference month. - MOF staff have access to preliminary fiscal data—within a week after month-end. These data are considered “soft” and are not available to other government agencies. - Internal access to fiscal data/GFS before general release is not officially made public, although it is a widely known fact. ### 1.2.3 Identification of statistical products - GFS tables the PRD provides to the SDS for dissemination do not include an explicit reference that the MOF is the source. - Accompanying text drafted by the SDS implies the MOF is the source of the data. - Elsewhere, the MOF is clearly identified as the source of the data. ### 1.2.4 Advance notice of methodological changes - Until now, there was no need for advance notice to the public of major changes in methodology, source data, and statistical techniques. - The MOF was not aware of the need for such notices as a good international practice. --- ### 1.3 Ethical Standards ### 1.3.1 Staff behavior guidelines - No specific code of conduct exists for the compilation of GFS. - All state employees are governed by a code of conduct according to the Law of Georgia on Public Service (No. 1022 of October 31, 1997), which includes penalties for violations. - This Law was disseminated to all staff when promulgated and is referenced in employment contracts. - The MOF is considering developing a code of conduct for its staff, similar to the Tax Department’s Code of Conduct. --- ### 2. Methodological Soundness ### 2.1 Concepts and definitions - Compilation of monthly and annual fiscal data follows the analytic framework prescribed by the annual budget laws. - Annual budget laws are generally consistent with the GFSM 1986 analytic framework but differ in significant respects: - Revenue and grants include repayments of loans previously extended by government. According to the GFSM 1986, these repayments are classified as Lending minus Repayments. - For local governments, proceeds from privatization are included as nontax revenue. - Lending minus repayments (“net lending”), which form part of total expenditure and net lending, are defined as amortization of foreign loans as are all disbursements of World Bank financed project loans. - As a result of the treatment of amortization of foreign loans, foreign financing is defined on a gross basis (i.e., new borrowing only). - In all data compiled by the PRD, including those for publication in the IMF’s GFS Yearbook, these budget law definitions are followed. - GFS compiled for some individual users follow an analytic framework that resembles GFSM 2001 in some respects: - Current expenditure include interest payments on a commitments basis. - Sales of equity other than shares (privatization) are recorded below the line as a disposal of a financial asset (financing). The 2002 Budget Law follows this approach. - The deficit/surplus is calculated on a commitments basis. - Financing and debt data are compiled only on a currency basis; foreign financing data exclude amortization of foreign loans. - There appear to be no official plans to migrate to the GFSM 2001 framework or methodology. ### 2.2 Scope - The scope of GFS disseminated (in terms of government units covered) generally meets GDDS recommendations; transactional coverage only broadly meets GDDS recommendations. - Monthly and annual GFS cover all activities of all general government units in Georgia, except: - Own revenues (e.g., fees and charges) of central and local government spending agencies received in their “Special Accounts” (and corresponding expenditure). - Starting with the 2002 budget, special accounts are included in the budget law, but monitoring them on a monthly basis is still problematic. Plans are to abolish special accounts in 2003. - Sufficient data on transactions between general government units are available to compile GFS for the consolidated general government, except transfers from the extrabudgetary funds to other general government units. - GFS do not cover any quasi-fiscal activities of government. - The PRD compiles nearly 20 different sets of tables for its various users; collectively these would cover most GFSM 1986 table requirements. - For the IMF’s GFS Yearbook, data are compiled for gross government debt (in accordance with international standards). - Government’s assumption of “historical loans” are not included in government debt; domestically, no data are disseminated on domestic government debt, and thus not total government debt. ### 2.3 Classification / sectorization - Institutional sectors of government are defined in accordance with the 1993 System of National Accounts (1993 SNA). - Except for data compiled for the IMF’s GFS Yearbook, GFS tables do not provide the same level of detail as required by international standards. - Revenue and grants: - Aggregate classification closely resembles GFSM 1986: distinction between tax and nontax revenue; tax revenues classified as taxes on income and profits, value-added tax, excises, customs duties, and other taxes; nontax revenues include repayments of government lending, property income, dividends from state enterprises, fines and penalties, and miscellaneous other nontax revenues. - Functional classification of expenditure and lending minus repayments follows COFOG in GFSM 1986. - Economic classification of expenditure and net lending resembles GFSM 1986 but differs in several ways: - All transfers to local governments are classified as current transfers, despite including transfers for capital purposes. - Loans to the Tbilisi Municipality are classified as current transfers; per GFSM 1986 these should be Lending minus Repayments. - A separate expenditure item, Targeted Programs, is included; according to international standards these should be classified to the appropriate economic categories. - All projects financed through the World Bank are classified as Lending minus Repayments, though most should be in appropriate current and capital expenditure categories. - Financing and debt data are classified by type of debt holder and debt instrument per GFSM 1986, except “changes in cash balances” are classified as “other liabilities” rather than Changes in Cash, Deposits, and Securities Held for Liquidity Purposes. ### 2.4 Basis for recording - Prices used to value flows and stocks reflect actual cash payments only: only cash transactions are recorded in the source data and thus GFS. - Debt liabilities are recorded at face value per GFSM 1986; amounts denominated in foreign currency are converted to Georgian Lari using the official exchange rate for the end of each month, as provided by the NBG. - Recording is done on a cash basis: revenue and expenditure data are based on cash received and paid; financing data are recorded on a cash basis per GFSM 1986. - For some purposes, financing data including foreign loans are shown, in line with interest, on a commitments basis (closer to GFSM 2001). - Grossing/netting procedures: - Government finance transactions are correctly shown on a gross basis. - Not all financing transactions are correctly recorded on a net basis. - Tax refunds are correctly classified as (negative) revenue transactions. --- ### 3. Accuracy and Reliability ### 3.1 Source data - Budgetary central government: - Timely and comprehensive revenue, grants and domestic financing data are available from the NBG Treasury Account records. - Timely data, but less comprehensive than revenue data, are available every month from the Treasury system. - Both revenue, grants and expenditure data are available on a cumulative year-to-date basis; no discrete monthly data are compiled in the source data. - Timely and comprehensive foreign financing data are available monthly from the MOF’s Debt Department. - Both domestic and foreign financing data are available on a monthly and a cumulative year-to-date basis. - No data are available on spending agencies’ operations relating to their own revenues received in their “Special Accounts.” - Starting with 2002, special accounts are included in the budget law, but monitoring them on a monthly basis is still problematic. Plans are to abolish special accounts in 2003. - Extrabudgetary Funds: - Timely, but aggregated data are available on revenues and grants, expenditures and financing from the two extrabudgetary funds. - Expenditure data do not identify transfers between them and other general government units to allow consolidation. - Revenues and grants of all general government units identify grants from other general government units separately, allowing consolidation. - Consolidated Central Government: - Timely data, but limited to the details of extrabudgetary funds’ data, are available from the Treasury Department. - No data are available on spending agencies’ operations relating to their own revenues received in their “Special Accounts.” - Local governments: - Timely cumulative year-to-date and annual data on local governments’ revenues, expenditures, and financing are available from the Budget Department’s Local Government Division. - Revenue data are quite comprehensive as NBG Treasury Account records are used, but exclude own revenues and some central government taxes collected by certain regions. - Local government expenditure data are very aggregated. - For final annual data, the same sources are used as verified by the Chamber of Control. ### 3.1.2 Fit of source data to required definitions and classifications - Source data broadly approximate GFSM 1986 definitions, classifications, valuation, and time of recording, with significant exceptions (see Classification/Sectorization). - The scope of source data is only broadly sufficient to analyze operations of the general government; not all government operations are covered. - Significant deviations: - Economic classification of expenditure in source data does not reflect international definitions. - Local government data do not allow economic classification of expenditure due to use of Soviet-era line-item classification; a broad functional classification is possible. - Data published on the outcome of the state budget are easily reconcilable with GFS data, as GFS are compiled to correspond exactly with the year’s budget law. - For some users, GFS compiled under a methodology that differs from the annual budget law are also easily reconcilable with annual outcome data. ### 3.1.3 Timeliness of source data - Source data for compiling GFS covering the general government and its subsectors are extremely timely: - Budgetary central government data (cumulative year-to-date) are available within seven days after the end of the reference month. Preliminary annual data are available by January 31 (year-end is December 31), and final annual data are available before April 1. - Extrabudgetary funds’ data (cumulative year-to-date) are available ten days after the end of the reference month. Preliminary annual data are available by January 31 (year-end is December 31), and final annual data are reported to the Treasury Department by March 1. - Aggregated Local governments’ data (cumulative year-to-date) are available 15 days after the end of the reference month. Preliminary annual data are available by January 31 (year-end is December 31), and final annual data are reported to the Budget Department’s Local Government Division by March 1. The consolidated data for all local governments are available by about mid-April. - Obtaining detailed local government data on a timely basis is problematic. ### 3.2 Statistical techniques - No statistical techniques are applied to compile GFS; no estimates are made for missing data (e.g., spending agencies’ “Special Accounts”). - Other statistical procedures (data adjustments, transformations, statistical analysis) are not employed in compiling GFS. ### 3.3 Assessment and validation of source data - Material gaps exist in coverage/completeness for some local government data, all government units’ own revenue sources and expenditures from these sources, and details of economic classification of expenditure of all general government units. - The MOF is attempting to address local government data issues but is constrained by existing laws governing central-local relations. - Detailed expenditure data are available in administrative records but are provided on an ad hoc basis to GFS compilers and users; no systematic efforts are made to improve availability of detailed central government source data. - There is an appropriate balance between accuracy and timeliness; tradeoffs are made between timeliness and level of detail. Preliminary data are replaced by final audited data annually. - The status of data (preliminary or final) is not always clearly identified in source data tables. According to the MOF: - All GFS compilers and users are aware that cumulative year-to-date data are preliminary. - Preliminary annual data are indicated as such; “if nothing is indicated” it means the annual data are final. - Data are corrected when revisions are made by the Treasury; such instances rarely occur. *Source: _cr03140 - 1.2 Transparency* ### 3.4 Assessment and Validation of Intermediate Data and Statistical Outputs ### 3.4 Assessment and Validation of Intermediate Data and Statistical Outputs ### Validation of intermediate data - Main intermediate GFS data are validated against original, unpublished source data (e.g., Treasury Department’s data provided to the PRD). - No validations of intermediate government finance data are carried out against other published sources, since no similar data are published elsewhere. - When GFS are compiled from different sources (e.g., NBG records for revenues rather than Tax Department records), discrepancies mainly arise due to different times of recording; compilers are aware of reasons and reconciliation is done as far as possible. ### Investigation of statistical discrepancies in intermediate data - 3.4.2: Not applicable. ### Investigation of discrepancies and indicators of problems in outputs - GFS are compiled from Treasury and NBG records and outputs are verified against original source data. - Discrepancies, which rarely occur, are assessed and investigated. - After compilation, all GFS data are compared with the previous period to identify unusual trends; explanations are obtained and documented internally. - Explanations for trends are not officially published but are provided to government users or available to other GFS users on request. ### 3.5 Revision Studies ### Routine studies and use of revision analysis - Two types of revisions to compiled GFS: - (i) ad hoc corrections by the Treasury Department to preliminary monthly data (e.g., due to more accurate information); - (ii) about four to five months after year-end, when preliminary data are replaced by final, audited data. - PRD compilers keep information on reasons for revisions to preliminary year-to-date data. - Reasons for differences between preliminary annual data and final audited data are provided in the Chamber of Control’s reports. ### 4 Serviceability ### 4.1 Relevance: monitoring and user needs - GFS use in Georgia is largely limited to users within government and international organizations. - Data timeliness is sufficient to assess whether government is achieving stated fiscal goals under the annual budget law framework. - Methodological basis and aggregated classification structure are generally not sufficient for analytical purposes: GFS are largely compiled on an aggregated basis and mainly for budget monitoring, which differs from international methodological standards. - No data on government debt are disseminated in the SDS and NBG publications (except foreign government debt disseminated domestically). - For sub-annual data, only cumulative year-to-date data are disseminated; no discrete monthly data are compiled. - MOF has no specific measures to obtain direct regular feedback from users; user needs are monitored indirectly via users’ initiative per General Administrative Code (No. 2181 of June 25, 1998) which requires agencies to provide information within ten days after request unless secret. ### 4.2 Timeliness and periodicity - 4.2.1 Timeliness follows dissemination standards: - Preliminary cumulative year-to-date GFS covering the general government (and its subsectors) are disseminated to the PRD’s list of users three weeks after month-end (some data are disseminated within government earlier). These data are disseminated in the SDS monthly bulletin and in the NBG monthly bulletin and website about two months after the end of the reference month. (GDDS recommendation: quarterly central government aggregates within one quarter after the end of the reference period.) - Preliminary annual data covering the general government (and its subsectors) are disseminated to the PRD’s list of users about one and one half months after the end of the reference month (some data are disseminated within government earlier). These data are disseminated in the SDS monthly bulletin and in the NBG monthly bulletin and website about three months after the end of the reference year. - Final annual data covering the general government (and its subsectors) are disseminated to the PRD’s list of users about five months after the end of the reference month (some data are disseminated within government earlier). These data are disseminated in the SDS Yearbook in the NBG monthly bulletin and website about six months after the end of the reference year. (GDDS recommendation: comprehensive annual central government data within six–nine months after the end of the reference period) - Monthly and annual data on government’s foreign debt are disseminated about a week after the end of the reference month in a local newspaper, and to other users within government and international organizations. Data on domestic government debt are only disseminated to users within government and to international organizations about two weeks after the end of the reference month. (GDDS recommendation: annual central government debt data within one–two quarters after the end of the reference period) - The timeliness of detailed local government data are sometimes problematic. - 4.2.2 Periodicity follows dissemination standards: - Cumulative year-to-date preliminary data for budgetary central government, extrabudgetary funds, and local governments are disseminated every month (cumulative year-to-date) and annually. (GDDS recommendation: quarterly and annual central government aggregates. Annual general government aggregates are encouraged by the GDDS.) - Central government foreign debt data are disseminated monthly and annually (GDDS recommendation: annual central government debt data, with quarterly data encouraged.) ### 4.3 Consistency - 4.3.1 Internal consistency: - GFS compiled by PRD from consistent sources (e.g., Treasury Department records on revenues, expenditure and financing) are internally consistent: deficit/surplus equals financing (with opposite sign); major aggregates equal sum of components; total expenditure and “net lending” by economic classification correspond with functional classification; domestic financing consistent with change in domestic debt (compiled from same NBG sources); foreign financing consistent with changes in foreign debt after accounting for amortization included in “net lending” and revaluations; transfers paid by one government level equal transfers received by another. - Lack of details in extrabudgetary funds’ expenditure data prevents verification or consolidation of transfers from these funds to other levels of government. - GFS compiled from separate data sources (e.g., Tax Department records vs. NBG bank account records) will show inconsistencies between deficit/surplus and financing due to checks received basis vs. checks cashed basis; statistical discrepancy equals amounts in transit. Inconsistencies also exist between expenditure data compiled from different sources. - Government foreign borrowing and amortization data are derived from UNCTAD DMFAS used by MOF. Budgetary central government financial assets data derived from NBG records of cash holdings. - 4.3.2 Reconciliability over time: - Compiled GFS follow annual budget laws; inconsistencies over time can be ascribed to changes in this law. - PRD does not apply material changes to classification or methodology to ensure a consistent series, but is aware of all such changes; reconciliation over a reasonable period could be done. - No official explanations of changes in trends are disseminated with the data, but PRD can explain them on inquiry. - 4.3.3 Consistency with other frameworks: - National accounts and balance of payments statistics are compiled from general government data provided by MOF to the SDS; monetary and financial statistics are compiled from NBG records. These data are consistent or reconcilable with GFS. ### 4.4 Revision Policy and Practice - 4.4.1 Schedule and visibility of revisions: - Ad hoc revisions to cumulative year-to-date data are done when needed; because only cumulative year-to-date data are published, revisions to previous months’ data are not visible to users. - Users of GFS published in SDS monthly bulletin or NBG bulletin are not informed of revisions to historical data. - Preliminary annual data are replaced by final audited data according to a regular, well-established and transparent schedule as required by law. - Internationally accepted practice (not followed) is to publish discrete monthly data in addition to cumulative year-to-date and to clearly identify revisions to historical data. - 4.4.2 Identification of preliminary data: - Preliminary data published in SDS and NBG bulletins are not always clearly identified. - Data provided by PRD to individual government and international organization users are identified as preliminary. - 4.4.3 Publication of revision studies: - Studies and analyses of revisions are not automatically made public. - Reasons for revisions are investigated and documented by PRD and are only made public when individual users enquire. ### 5 Accessibility ### 5.1 Data accessibility - 5.1.1 Presentation for interpretation and comparison: - Presentation of disseminated GFS is generally not adequate for proper interpretation and meaningful comparisons: only major aggregates and balancing items are published; detailed classification tables set out in GFSM 1986 are not covered. - Annual time series data are provided only for a short period. - GFS disseminated to individual users are more detailed per request. - Only data on foreign government debt are disseminated domestically. - 5.1.2 Dissemination media and formats: - No dedicated GFS publication or website. - SDS and NBG dissemination formats are generally adequate, but only for aggregated GFS data. - GFS are disseminated to individual users as requested, usually by facsimile or e-mail. - 5.1.3 Pre-announced schedule: - Dissemination policy for GFS is not publicized; no advance release calendar exists; actual date of publication may vary. - 5.1.4 Equal access to data: - GFS are not made available to all users at the same time. - Users within government and international organizations have access as soon as data are compiled. - In principle, any user who requests data at that time could obtain them; preferential internal access is not publicly declared. - 5.1.5 Availability of nonpublished sub-aggregates: - Unpublished but nonconfidential data are made available upon request at no fee under General Administration Code provisions; availability and terms are not publicized in GFS publications. ### 5.2 Metadata accessibility - 5.2.1 Documentation on concepts and recording: - No separate metadata exist on concepts, scope, classifications, basis of recording, data sources, and statistical techniques for GFS. - Budget Systems Law and annual budget laws contain some information on concepts, scope, definitions, and basis of recording; GFS currently compiled follow these laws. - Differences between internationally accepted standards and GFS compiled in Georgia are not annotated nor publicized. - 5.2.2 Levels of detail adapted to audience: - Apart from Budget Systems Law instructions, no metadata exist on sources, methods, and methodology for GFS compilation. ### 5.3 Assistance to users - 5.3.1 Contact person: - No contact person for GFS is publicized in SDS or NBG publications. - Individual users who receive GFS from PRD know who to contact; otherwise users contact PRD or any MOF department. - 5.3.2 Catalogues and charges: - MOF does not have catalogues of publications, documents, or services. ### Key assessments from Table 4 (selected) - 0.1 Legal and institutional environment: Assessment symbol X; comment: Two departments in MOF assigned responsibility for compiling GFS after recent restructuring posing risk of duplication; legal mandate/measures not fully adequate to ensure timely reporting of detailed data by all local governments. - 0.2 Resources: Assessment symbol X; comment: Staff and financial resources mostly insufficient for comprehensive GFS compilation; no resources for migration to GFSM 2001. - 0.3 Quality Awareness: Assessment symbol X; comment: GFS quality control processes not formally documented; short notice for data compilation may compromise quality; tradeoffs between quality, timeliness, and degree of detail not communicated to users. - 1.1 Professionalism: Assessment symbol X; comment: Constrained by annual budget laws and insufficient recognition of user needs for detailed GFS that correspond to international standards. - 1.2 Transparency: Assessment symbol X; comment: Terms and conditions under which GFS are compiled and approval process for publication are not public; MOF as compiler not always clearly identified. - 2.1 Concepts and definitions: Assessment symbol X (Practice Largely Not Observed); comment: Analytic framework generally GFSM 1986, but revenue, expenditure, and financing concepts & definitions not according to GFSM 1986; expenditure definition includes debt amortization; no official plans to migrate to GFSM 2001. - 2.2 Scope: Assessment symbol X; comment: GFS cover all general government units, but not all their transactions—own revenues and expenditures from these resources are excluded. - 2.3 Classification/Sectorization: Assessment symbol X (Practice Not Observed); comment: Classifications broadly according to GFSM 1986 except lacking detail; economic classification of expenditure includes nonstandard items and classifies some current and capital expenditures and amortization of foreign financing as lending minus repayments. *Source: _cr03140 - 3.4 Assessment and Validation of Intermediate Data and Statistical Outputs* ### 3. Accuracy and reliability ### _cr03140 - 3. Accuracy and reliability ### 3.1 Source data - Generally adequate, but no data up to now on spending agencies’ activities regarding own revenue sources; lack of details for extra-budgetary funds expenditure do not allow full consolidation of data; levels of detail and classifications of local government expenditures do not approximate classification required for GFS; lack of details to compile complete economic classification of expenditure (e.g., no breakdown readily available of subsidies and other current transfers). ### 3.2 Statistical techniques - - ### 3.3 Assessment and validation of source data - Gaps in the coverage/completeness of the data collection are assessed, but not much could be done to improve because of laws and/or lack of resources. ### 3.4 Assessment and validation of intermediate data and statistical outputs - - ### 3.5 Revision studies - - ### 4. Serviceability #### 4.1 Relevance - Methodological basis and classification structure generally not suitable to meet users’ analytical needs. - No measures in place to formally seek feedback from users on relevance and usefulness of statistics—feedback is obtained indirectly through individual user requests. #### 4.2 Timeliness and periodicity - - #### 4.3 Consistency - Data internally consistent. - No consistent GFS time series exist because of budget law changes but it is feasible to compile such series; no explanations provided for inconsistencies, but aware of reasons, so can reconcile. #### 4.4 Revision policy and practice - Only cumulative year-to-date data published every month, so revisions are not visible; preliminary data sometimes not clearly identified, studies and analysis of revisions only made public in response to individual user requests. ### 5. Accessibility #### 5.1 Data accessibility - Presentation of aggregated GFS in SDS and NBG publications not adequate to allow proper interpretation and meaningful comparisons; GFS not released on a pre-announced schedule; GFS not made available to users at the same time; nonpublished but nonconfidential data are made available to users upon request, but this is not publicized. #### 5.2 Metadata accessibility - No metadata exist on GFS. #### 5.3 Assistance to users - Contact person for each subject field not publicized. *Source: _cr03140 - 3. Accuracy and reliability (PDF chapter/section).* --- ### Monetary Statistics — Selected Quality and Institutional Findings ### 0. Prerequisites of Quality — Legal and institutional environment - Article 56 of the Organic Law on the NBG of 1995 gives the NBG authority for “collecting statistical reports, accounting data and other materials from banks and other financial institutions, other government agencies, and private companies and organizations.” - Commercial Banks prepare and submit reports as prescribed by the Law of Georgia on activities of Commercial Banks (LCB) of 1996, Article 29. - Article 68 of the Organic Law provides that the NBG may publish financial statements, reports, and studies on financial and economic issues as it shall deem appropriate. - The Macroeconomic Research and Monetary Policy Department (MRMPD) is responsible for research on statistical methodology, data collection, processing, and dissemination of monetary and financial statistics (MFS). Banks’ statistical data are collected by the Monetary Statistics and Information Division (MSID). - Within the NBG there are arrangements to ensure information flow between the AD, BSD, MOD, FED, and MSID; there is no centralized database and data are shared among departments through e-mail. - The MPD compiles analytical accounts for domestic dissemination and to report to the European II Department of the IMF, while the MSID compiles similar but more detailed information to report to the Statistics Department of the IMF — this arrangement does not encourage resolution of methodological differences between the two sets of data. ### 0.1.3 Confidentiality and access - Article 21 of the NBG’s Organic Law forbids publication or disclosure of confidential information (paragraphs 1–3). - Only authorized staff members have access to individual institutions’ data before aggregation; procedures prevent individual disclosure; computers are password protected. ### 0.1.4 Reporting mandates and enforcement - Article 56 provides legal basis for statistical collection; NBG may request information and investigate accuracy. - Circulars/guidelines require reporting forms to be submitted no later than five working days after the end of the reporting period (LCB Article 29 paragraph 1). - Article 30 of the LCB: failure to comply is punishable by a fine and the NBG may suspend or restrict institutions’ operations in the event of nonresponse. ### 0.2 Resources #### 0.2.1 Staff, financial, and computing resources - The MPD has a total of five staff, two of whom focus on compilation of monetary statistics. - The MSID has a total of eight staff, who are professionals exclusively devoted to the compilation of statistics. - Three people (in MSID) focus on statistical concepts and methodology and the improvement of coverage and accuracy of MFS. - Staff of the MSID and two specialists of the MPD devoted to the compilation of MFS have university degrees in economics or statistics. The head of the MSID holds a Ph.D. in economics. - Reporting commercial banks submit returns through e-mail and data are entered into a compilation system that performs validation checks. - There are currently no budgetary constraints impeding data collection and compilation activities. #### 0.2.2 Measures for efficient use of resources - MRMPD management holds meetings with staff to enhance policy vision and understanding. - NBG budgeting is annual by the Board every December; departments identify needs and the Board allocates budgets. - The AD measures cost-effectiveness quarterly; reviews are conducted regularly; new technology for data processing and dissemination is tested by computer and system analysis experts. ### 0.3 Quality Awareness #### 0.3.1 Processes to focus on quality - The NBG exercises quality control at every stage of data production and dissemination, although no explicit measures are undertaken to focus on quality overall. - MSID verifies reporting practices, has systems and procedures to ensure quality; all levels of MSID participate in data review prior to publication. - Source data are crosschecked for accuracy; discrepancies are investigated; validation procedures are both automatic and visual; MSID consults reporting institutions to verify misclassifications. - An external auditing company audits the financial statements of the NBG. #### 0.3.2 Monitoring quality of collection, processing, dissemination - Quality is monitored through crosschecks; reporting institutions are informed and guided for collection problems. - No formal surveys of users; users can contact MSID directly or through NBG’s website and responses are provided to all requests. - No other body outside MRMPD provides guidance on quality of monetary statistics. #### 0.3.3 Dealing with quality tradeoffs and planning for emerging needs - MSID monitors developments in financial markets and analyzes inclusion of nonbank depository corporations and other financial institutions (insurance companies and pension funds) into monetary statistics. - There is recognition of tradeoffs between data quality and timeliness. - Production of monetary statistics is expected to be fully automated in the nearest future; a new system for monetary data compilation and processing is in a testing stage and will include checks and validations at every stage. - To accommodate reporting delays, some data may be published as provisional. - Informal meetings are held periodically with policy makers and other data users; NBG invites user comments via its website. ### 1. Integrity #### 1.1 Professionalism - Article 3 (paragraph 1) of the NBG Law stipulates that the NBG shall be independent in its operations. - Professionalism promoted by participation in lectures, conferences, and meetings with other professional compilers. - Compilation is based on balance sheets of reporting institutions; the depository corporations survey includes the NBG and all commercial banks, reducing scope for altering results. - NBG and commercial banks utilize a common chart of accounts that conforms to international accounting standards. - The MSID comments on erroneous interpretation of its statistics if deemed necessary and provides explanatory notes in publications and website. #### 1.2 Transparency - The NBG Law is available on the NBG website and in hardcopy on request. - Statistical publications provide e-mail addresses and telephone numbers for additional information. - Following MRMPD approval, a limited number of senior NBG staff have access to data prior to publication. - Governmental agencies (Financial-Budgetary Committee of the Georgian Parliament, Ministry of Finance, State Department for Statistics) have access to monetary statistics prior to public release; internal access is not made public in terms of who has access and when. - MSID is identified as the source of monetary statistics published in the Monetary and Banking Statistics Bulletin (MBSB), published quarterly providing monthly information. - Changes in data classification, source data, and statistical techniques are announced simultaneously with the release of the data; explanation is provided as a note in the publication where the change first takes place. Advance notice is given only in the case of major changes. #### 1.3 Ethical standards - The Code of Conduct for the Staff of the NBG provides broad ethical guidelines; new staff members are informed about the guidelines. ### 2. Methodological Soundness #### 2.1 Concepts and definitions - The analytical framework used by the NBG reflects concepts and principles based on the IMF’s Monetary and Financial Statistics Manual (MFSM). - The depository corporations survey consolidates accounts of the central bank and other depository corporations to present intermediation roles. - Monetary and credit aggregates in the depository corporations survey: - (i) money supply M1 (currency in circulation and demand deposits); - (ii) M2 (M1 plus time and savings deposits); - (iii) broad money or M3 (M2 plus residents’ foreign exchange deposits). - Other aggregates identified: - (i) foreign assets (net); - (ii) domestic credit (disaggregated by claims on central government (net), claims on local governments, claims on public nonfinancial corporations, claims on other resident sectors and claims on other financial corporations); - (iii) other items (net). *Source: _cr03140 - 3. Accuracy and reliability (PDF chapter/section).* ### 2.2 Scope ### 2.2 Scope ### 2.2.1 Scope overview - The depository corporations sector of Georgia comprises 25 commercial banks and two branches/subsidiaries of foreign banks operating in Georgia. - According to the LCB of 1996, “Bank” means a legal entity licensed by the NBG that attracts deposits and uses them on its own behalf to conduct banking activities pursuant to applicable law. - Institutional coverage of monetary statistics includes: - banks’ domestic headquarters and their domestic branches, and domestic branches of foreign banks operating in Georgia. - It does not include other depository corporations that are not banks according to the previous definition. - Data on nonbank depository corporations will be collected by the NBG in the near future through a department recently created within the NBG to supervise their activities. - Currently, the size of these institutions is not significant if compared to the commercial banks. - Data on other financial corporations, such as insurance companies and pension funds, which are considered to be significant, are not covered in a financial corporations survey. ### 2.3 Classification/Sectorization - The NBG uses the residency criterion to distinguish between domestic and external accounts for the compilation of monetary statistics; the delineation is broadly consistent with BPM5, 1993 SNA, and MFSM. - The sectoral classification recommended in the MFSM for compiling monetary statistics refers mainly to the following subsectors: - (i) central bank; - (ii) other depository corporations; - (iii) other financial corporations; - (iv) central government; - (v) state and local government; - (vi) public nonfinancial corporations; and - (vii) other resident sectors. - In NBG monetary statistics the following sectoral groupings of resident institutional units are distinguished: - (i) central government; - (ii) local governments; - (iii) central bank; - (iv) depository corporations (no public banks); - (v) foreign depository corporations; - (vi) other nonbank financial institutions; - (vii) nonfinancial public enterprises; and - (viii) other sectors. - Limitations noted: - Some financial instruments are not fully sectorized owing to their relatively small size (e.g., securities other than shares and other equity). - Some instruments are not identified at all (e.g., financial derivatives). - Estimation of accrual interest by counterpart sector for some financial instruments is not carried out where information is not available. ### 2.4 Basis for Recording #### 2.4.1 Valuation of flows and stocks - MFSM general recommendation: valuation on the basis of market prices or market-price equivalents (fair values), with loans as an exception—loan values should be based on creditors’ outstanding claims without adjustment for expected loan losses (book value = outstanding principal plus any accrued interest). - NBG financial statements are prepared in accordance with related Georgian Legislation and requirements of the NBG Law, which is based on international accounting standards since 2001. - Depository corporations record securities as follows: - (i) securities held to maturity are recorded at face value with discount/premium shown on the counter account; - (ii) securities held for trading—at market prices; and - (iii) securities held for sale—the same way as securities held to maturity but at the same time subject to revaluation. - Consistent with MFSM recommendations: - Loan portfolio and deposits on banks’ balance sheets are valued at book value. - Loan valuation is not adjusted for expected loan losses. - Provisions for expected loan losses are recorded as separate entries on the liability side of the balance sheet. - Foreign-currency stocks and flows: - MFSM: convert to national currency at market exchange rate prevailing at time of entry. - In Georgia: foreign assets and liabilities are recorded at the official exchange rate set at the Tbilisi Interbank Currency Exchange on a daily basis. - Flows components: - MFSM recommends compiling stocks and each of transactions, revaluations, and other changes in volume. - In Georgia, banks provide financial information on transactions for deposits and loans (largest instruments). - Securities other than shares accounted for only 4.3 percent of total assets in April 2002. - The NBG does not derive transactions for monetary statistics from stock data. #### 2.4.2 Recording on an accruals basis - MFSM accrual principle: interest due but not paid should be incorporated into the outstanding amount of the financial asset or liability. - Georgian accounting standards require interest to be recorded on a monthly basis; interest arrears are recorded separately when past due. - Sectoral breakdown of accrued interest is only partially sectorized: - Accrued interest on loans and advances to, and deposits of, the general government, public nonfinancial corporations, other depository corporations and other residents are separately identifiable from the chart of accounts. - Values of accrued interest for republican government, local government, resident and nonresident banks, other financial corporations, and private sector are not separately identified. - Accrued interest should be distinguished by instrument and sector. - Transactions are generally recorded at the time they occur. #### 2.4.3 Grossing/netting procedures - Assets and liabilities of the NBG, banks, and other financial institutions are collected and compiled on a gross basis, consistent with MFSM. - Claims on particular transactors are not netted against liabilities to those transactors. - Data presented on a net basis (e.g., “Foreign assets (net)” and “Claims on central government (net)”) are also shown with the underlying gross data. - For sectoral balance sheets of the NBG and depository corporations, financial assets and liabilities are aggregated into major categories (e.g., claims by debtors, deposits by creditors). - The depository corporations survey cancels out outstanding claims and liabilities between the NBG and banks and between the banks. *Source: _cr03140 - 2.2 Scope* ### 5.2 Metadata Accessibility ### _cr03140 - 5.2 Metadata Accessibility ### Metadata accessibility — monetary statistics - The monetary statistics’ metadata documentation, including the concepts, classifications, data sources, characteristics, and definitions is not available. - Development of the metadata documentation is under way. - Draft metadata are being prepared in accordance with IMF recommendations for GDDS participating countries. - Assessment summary entries: - In the Data Quality Assessment Framework table, 5.2 Metadata accessibility is assessed as "NO" with the comment: "No documentation on metadata exists." - The dissemination format (Acrobat files) is judged "not user friendly" and there is "no advanced release calendar." ### Assistance to users (monetary statistics) - Service and contact information: - Prompt and knowledgeable service and support is provided to statistics users, but not all statistical releases identify specific individuals. - Phone numbers and e-mail addresses of the NBG’s heads of departments and other specialists can be found in the “Contacts” section of the NBG’s website. - Names of people responsible for the preparation of published statistics, as well as contact e-mails, are cited in the MBSB. - Publications and charging: - Main statistical publications, documents, and other services are free of charge. - A catalogue of publications is not available in hardcopy; the NBG’s website lists publications available to the public. - Subscriptions are handled by the MSID. - Periodic publications are disseminated on the website of the NBG. - Subscription to hard copies for certain publications such as the journal Bank are not free. - User outreach gaps: - No brochures have been developed to educate users. - No formal user surveys are undertaken to obtain feedback on data quality issues. ### Data Quality Assessment Framework — monetary statistics (selected findings) - Organizational/quality prerequisites: - Compilation of monetary and financial statistics for surveillance and publication is conducted separately by two divisions within the NBG. - Nonbank depository corporations are not covered by MFS and are considered "not very significant." Plan: "Collect information on the nonbank depository corporations by March 2003." - Integrity and methodological soundness: - Accrued interest not sectorized. - Draft metadata being prepared, but "No documentation on metadata exists." - Serviceability and accessibility: - Lack of established process of consultation with users. - Revision studies are "seldom carried out." - Dissemination format (Acrobat files) not user friendly; no advanced release calendar. ### Balance of payments statistics — governance and legal framework - Legal and institutional basis: - State Department of Statistics (SDS) operates under the Law on Statistics No. 1071 of November 12, 1997, with an amendment in June 1999. - Regulations for the SDS No. 92 of February 28, 1998, and the State Program for Development of Statistics for 2001–2005 explicitly mention compilation and dissemination of balance of payments statistics. - Article 11 of the Law guarantees the SDS access to data produced by legal entities, including data produced by government ministries and the NBG. - Organizational challenges: - SDS has headquarters, thirteen regional offices, and several local offices; regional heads may be appointed with regional influence, creating dual allegiance. - Dual reporting lines can result in lower priority for national statistics (such as balance of payments) at the regional level. - Weak communication and distance from headquarters may adversely affect survey administration and data collection. ### Balance of payments statistics — data sharing and sources - Interagency arrangements: - Written agreements exist between the SDS and the MOF, the NBG, and the SDC for sharing macroeconomic data. - The agreement with the SDS requires only that the SDC provide a computer file of its trade statistics database, with no requirement for SDC quality control or adherence to international standards. - As a consequence, the SDC database has major gaps in coverage, valuation, and classification and requires the BOPD to spend resources checking for errors and inconsistencies. - Duplication and coordination: - Apart from internal BOP estimates by the NBG, there is no evidence of major duplication of statistical effort among agencies. - No regular contacts with other statistics producing agencies are maintained to take into account respondent burden. ### Balance of payments statistics — confidentiality and reporting - Confidentiality: - Confidentiality of individual data reported to the SDS is guaranteed under Articles 7, 12, and 17 of the Law. - The Law does not provide penalties for disclosure by staff; SDS is planning an amendment to include penalties. - Anti-Monopoly Law provisions allow the Anti-Monopoly Department to request confidential data from the SDS, creating a legal conflict that is unresolved. In practice, no breaches of confidentiality of individual balance of payments data have occurred. - Staff sign a declaration on joining the SDS confirming awareness of confidentiality requirements; access to individual data is restricted to appropriate staff and senior staff review data before dissemination. - Legal mandate and survey response: - Article 11 authorizes the SDS to request information from all legal public and private institutions and individuals; the June 1999 amendment provides some penalties for nonreporting but these are not considered effective for survey nonresponse. - Few statistical surveys of balance of payments are undertaken (currently confined to a survey of transportation services and a survey of inward FDI). - BOPD has little experience conducting surveys and has been reluctant to use legal authority to mandate respondents; voluntary surveys have seen poor response. ### Resources, staffing, and IT capacity (BOPD) - Staffing and turnover: - There are eight staff members in the BOPD. - The BOPD has experienced heavy staff turnover: nine different heads of the BOPD in the last seven years (acting or permanent). - Inadequate remuneration is a major constraint on retention; SDS salaries have fallen to about 7 percent of the 1998 levels in real terms. - IT resources: - The BOPD has five computers, of which only two are of functional value. - Lack of computers is aggravated by disproportionate use of computer time to service user requests, especially for trade data. - Resource management: - Internal processes exist to measure resources used to compile statistics and periodic reviews of budgeting procedures are undertaken. - Failure to use the SDS legal mandate to conduct mandatory balance of payments surveys has led to inefficient use of resources by attempting surveys on a voluntary basis. ### Quality awareness and quality management (BOPD) - Institutional initiatives: - SDS includes a Statistical Research Institute (methodology and software) and an Institute for Informatics and Telecommunications (information and analytical systems). - A Scientific and Advisory Council established in 1997 includes external specialists and provides assessments and recommendations. - Shortcomings in quality processes: - These initiatives have not substantially contributed to quality awareness for balance of payments statistics. - The BOPD has identified collection-stage problems but has not resolved major issues such as voluntary survey nonresponse. - Processes are not established for identification and resolution of problems during processing and dissemination. - Since 1998 SDS has held meetings with data users; user surveys indicate widespread dissatisfaction with the quality of balance of payments statistics. - Trade-offs and planning: - Senior management recognizes trade-offs between timeliness and accuracy, but limited monthly and quarterly data sources mean the trade-off is not a major current issue. - Introduction of comprehensive annual surveys would necessitate addressing tradeoffs and reallocating resources to improve annual data quality. ### Integrity, transparency, and professional independence (BOPD/SDS) - Professionalism and methodological independence: - The Law on Statistics underscores professional independence of the SDS. - SDS is independent in choosing sources and methods; no instance of ministerial interference is reported. - SDS provides press briefings, technical commentary, and monitors media for misuse or misinterpretation of data. - Transparency shortfalls: - The SDS web page reproduces information on terms and conditions for data collection and compilation and provides summary confidentiality information, but statistics publications identify information about the SDS and not about specific products. - The survey forms indicate only some terms and conditions (e.g., confidentiality) and do not fully spell out all terms and conditions. - There is no public information on who has access to data prior to release; once data are finalized they appear to be sent to government and other privileged users before public release, but this practice is not documented or published. - SDS products are clearly identified by name and logo; the NBG tables identify the SDS as source, but the SDS does not request users to attribute statistics to it when reproduced. - Users are not given advance notice of major changes in methodology, source data, or statistical techniques. *Source: _cr03140 - 5.2 Metadata Accessibility (Compiling Agency: National Bank of Georgia; State Department of Statistics), extracted from the provided PDF chapter.* ### 1.3 Ethical Standards ### 1.3 Ethical Standards ### Ethical rules and staff awareness - Ethical rules are laid down in the Code of Conduct for Civil Servants. - New staff are made aware of the Code and the mission of the SDS when they join the department. ### Methodological Soundness — Concepts and Definitions - Balance of payment statistics are compiled broadly in accordance with the guidelines set out in the IMF’s BPM5. ### Methodological Soundness — Scope - With one exception, the scope of balance of payments statistics is fully consistent with internationally accepted standards. - Exception: the exclusion of reinvested earnings from the definition of direct investment income. - The balance of payments covers the economic territory of the country inclusive of Abkhazia and the South Ossetia Regions, and the interpretation of residence of entities and individuals follows BPM5. ### Methodological Soundness — Classification/Sectorization - Classification broadly in accordance with BPM5. - Current account classification: goods, services, income, and current transfers. - Goods: sub-classification does not separately identify goods for processing, repairs on goods, and nonmonetary gold. - Transportation services: transportation of goods (including transit trade), passengers, and ancillary transportation services. - SDS classification of other services: include travel, communications, freight insurance, business services, and government services; exclude construction, nonfreight insurance, financial, computer and information, royalties and license fees, personal, cultural, and recreational services. - Income: compensation of employees, direct investment income, portfolio income, and other investment income, each identified separately. - Current transfers: comprise, without separate identification, grants in cash and goods received by the government as part of official development assistance from foreign governments, workers’ remittances, withholding taxes, and contributions to international organizations. - Capital account: comprises migrants’ transfers. Grants made in cash and goods as part of official development assistance for expenditures of a capital nature are not identified separately and are included with current transfers. - Financial account: direct investment, portfolio investment, other investment, and reserve assets. - Direct investment: transactions identified separately for direct investment in Georgia and abroad; no subclassification by instrument (equity capital, reinvested earnings, other capital). - Portfolio investment: classified only broadly into assets and liabilities without further breakdown by instrument or sector. - Other investments: shown separately for assets and liabilities with subclassifications for instrument or sector. - Reserve assets: monetary gold, special drawing rights, reserve position in the IMF, and foreign assets in the form of currency, deposits, and securities. - Four sectors identified: the monetary authorities, general government, banks, and other sectors. Other sectors comprise without separate identification, nonbank financial institutions, enterprises, and individuals. ### Methodological Soundness — Basis for Recording - Transactions are valued at market prices. - Data collected either in Georgian Lari or the U.S. Dollar; currency conversion when necessary takes place at monthly or quarterly period average exchange rates according to the periodicity of the data sources. Actual rates are used when available. - When flows are derived from stock positions, adjustments are made to remove the effects of exchange rate fluctuations. - In some instances, the rules of currency conversion have not followed BPM5 guidelines. - Transactions are generally recorded on an accrual basis. - Grossing/netting procedures: current account items recorded on a gross basis; financial account items recorded on a net basis, separately for individual asset and liability components. ### Accuracy and Reliability — Source Data: overview - The data sources available for balance of payments compilation are inadequate. - Potential data sources are not kept under continuous review, and very little attempt is made to monitor current balance of payments developments. - Only very recently the BOPD has started monitoring the financial press for information on international transactions. - Data sources comprise mainly administrative data and statistical surveys (a survey of transportation services and a survey of inward FDI). ### Accuracy and Reliability — Source Data: category-specific sources and limitations - Goods: - Trade data provided by the SDC based on the Single Administrative Document (SAD). - SAD includes information on currency and exchange rates, mode and terms of transportation, and valuation based on both c.i.f. and f.o.b. basis. - SDC does not collect data on goods for processing, repairs on goods, procurement of goods in ports by carriers, nonmonetary gold, or imports of gas, and imports and exports of electricity. - SDC has not fulfilled commitments to SDS to provide information and revised trade data on smuggled goods (cigarettes, oil products, alcoholic beverages, and other consumer commodities). - Transportation services: - Transportation credits derived from survey data and data supplied by pipeline companies. - Survey limited to public enterprises; private sector vehicle operators not identified; some estimates made but not considered of good quality. - Earnings from transit trade: estimates from railways, pipeline and oil companies are of good quality. - Data for other transportation services are incomplete. - Travel services: - Travel credits estimated from number of tourist arrivals and estimates of average expenditure of tourists. - Average expenditure estimated from interviews at airports by the BOPD, TACIS-assisted sample survey carried out in 1999 and 2000, and survey data. - All principal airports, seaports, railway stations, and border crossing points in Georgia were covered. - Travel debits calculated from data on travel abroad by residents of Georgia. - Communication services: data from the Ministry of Communications. - Insurance services: no data sources developed. - Other business services: limited data from FDI enterprises. - Government services n.i.e.: data from MOF sources and a survey of foreign embassies in Georgia conducted in 1997. - Other services (nonfreight insurance, construction services, royalties and license fees, financial services, computer and information services, personal, cultural, and recreational services): no data sources developed. - Compensation of employees: only data source is numbers of short-term workers abroad, broken down by host country. - Direct investment income: debits estimated from some limited data sources provided by FDI enterprises. - Portfolio investment income: source data are banking returns sent to the NBG. - Other investment income: - Interest on government debt: External Debt Management Unit of the Ministry of Finance. - Interest arrears shown as interest payments with contra entries in the financial account as short-term liabilities. - Interest on reserves: data provided by the NBG. - Current transfers: - Receipts: official development assistance (ODA), workers’ remittances and expenditures of nongovernment organizations (NGOs). - ODA data from the OECD database. - Workers’ remittances calculated on the basis of migration patterns and data from the household surveys. - Expenditures of NGOs estimated based on their number and nature of activities in Georgia. - Debits: taxes paid to foreign governments by Georgia’s short-term workers derived from the model used to estimate compensation of short-term workers abroad. - Debits attributed to workers’ remittances: earnings sent abroad for family support by expatriate workers including long-term technical personnel. - Capital transfers: identified only for migrants’ transfers. - Direct investment: currently no data sources for inward direct investment. The SDS survey of direct investment enterprises yielded poor results because of nonresponse. - Portfolio investment: data comprise portfolio investment by banks obtained from the NBG. - Other investment: - Bilateral and multilateral loans data from the Debt Management Unit in the MOF; data on disbursements, principal repayments, interest payments, and arrears are available regularly on a quarterly basis or when needed by the SDS. - Information on loan agreements and rescheduling available from the MOF unit. - Changes in foreign assets and liabilities of commercial banks from the NBG based on commercial bank returns; adjustments made to remove effects of revaluations due to exchange rate movements. - Estimates for trade credit liabilities based on the value of imports. - No data included for transactions of insurance companies and pension funds. - Reserve assets: data from the NBG consist of monetary gold, holdings of special drawing rights, reserve position in the IMF, and foreign exchange and other claims. Valuation changes are excluded in the flow data submitted by the NBG. - Although the Abkhasia and South Ossetia Regions are included in the economic territory of Georgia, almost no source data are collected on balance of payments transactions for these regions. ### Accuracy and Reliability — Source Data: assessment of fit to definitions and procedures - BOPD uses procedures to improve coverage, valuation, and classification, but procedures are not always appropriate and, for many deficient source data, no procedures have been developed. - Goods: BOPD makes a global adjustment to Customs data using a simple grossing up ratio without objective basis. - Trade data reconciliation with Armenia, Azerbaijan, and Turkey for 1998–99 was initiated but not completed. - Attempts to adjust Customs data using partner country data from IMF’s Direction of Trade Statistics were started but not completed. - No estimates made for smuggled imports/exports despite special survey results (e.g., Household Budget Survey on tobacco consumption for 1998) not being used to adjust Customs data for cigarette imports. - Transportation: freight debits estimated by a rule of thumb of five percent of total value of imports c.i.f.; no sample-based estimation of c.i.f./f.o.b. differences from SAD records. - Insurance: rule of thumb estimate of two percent for insurance on freight; no use of SAD records or contact with insurance companies; no reporting by insurance companies of their balance of payments transactions. - Travel: TACIS-assisted surveys provided useful data but were both carried out in the fourth quarter of consecutive years, preventing analysis of seasonal effects. - Transfers: ODA received in form of capital goods are not identified; no estimation from Customs or MOF records or project information. - Direct investment: media sources not effectively used to monitor major direct investment activities. ### Accuracy and Reliability — Timeliness - Administrative data from ministries and agencies are timely. - Direct investment surveys have experienced a high nonresponse rate despite deadlines and follow-up procedures. - OECD data on transfers are not timely. ### Accuracy and Reliability — Statistical Technique and Procedures - BOPD does not conduct large statistical surveys; computer processing not utilized to edit and tabulate questionnaires. - Sound statistical procedures are not used for data adjustments and transformations. - The c.i.f./f.o.b. factor for valuing imports is based on a rule of thumb of five percent of the value of imports and is not estimated from a sample survey of transportation enterprises. - Estimates of shuttle imports are not based on periodic surveys at major border points or extrapolated using monthly volume of shuttle traders identified by customs and immigration authorities. - Adjustments for underestimation of imports and exports based on partner country data are not specifically made; part of a global estimate for under coverage and under valuation. ### Accuracy and Reliability — Assessment and Validation of Source and Intermediate Data - Verification of relational data and trend analysis to detect inconsistencies are not routinely performed. - Very little monitoring of the media for information on international transactions. - Behavior of series is usually not crosschecked with related series or indicators. - Investment income payments and receipts not regularly assessed in relation to corresponding stock data in the international investment position. - Freight earnings not regularly assessed in relation to value/volume of trade flows. - Financial data not reconciled with changes in corresponding stock data for external debt and other elements of the international investment position. - Travel-related transactions analyzed in relation to immigration data on international travelers. - Net errors and omissions are not analyzed to monitor developments in their size or sign. - Bilateral trade comparisons have been done with major partners, but follow-up actions to resolve discrepancies were not taken. - External debt data are not routinely compared with the Joint BIS-IMF-OECD-World Bank Statistics on External Debt or the BIS locational international banking statistics. ### Accuracy and Reliability — Revision Studies - Direction and magnitude of revisions between preliminary and final data are not investigated. - Studies on fluctuations in the data, including sources of errors and omissions, are not carried out routinely. ### Serviceability — Relevance and User Engagement - Specific actions are not taken to ensure balance of payments statistics continue to address users’ needs. - No established process of consultation with users in government or private sector. - Feedback is received from the IMF regularly and informally from the NBG; no established process of periodic review or users’ surveys to assess relevance. - SDS regularly participates in international statistical meetings and seminars. - Detailed classification and sectorization recommended in BPM5 are available to users only on request and are not made public; only a very summary classification and sectorization is provided in SDS publications. *Source: _cr03140 - 1.3 Ethical Standards (Balance of Payments Statistics).* ### 4.2 Timeliness and periodicity ### _cr03140 - 4.2 Timeliness and periodicity ### Timeliness and periodicity - Annual balance of payments data are disseminated within four months after the end of the reference period. - Quarterly balance of payments data are disseminated within a quarter after the end of the reference period. - Additional publication lags may occur; these would be remedied by the use of the SDS website for the release of data. - The SDS compiles quarterly and annual balance of payments statistics, thereby exceeding the GDDS requirements. - In practice, finalized quarterly data are considered released "as soon as the data are finalized (usually 45 days after the end of the reference period for the quarterly and four months after the end of the reference period for the annual data)" and are available on request to government bodies and selected agencies. - The SDS plans to publish balance of payments data on its website. ### Consistency - Concepts, definitions, and classifications for quarterly balance of payments data are the same as for annual data; annual data are derived from quarterly data and are not subsequently revised. - Net errors and omissions in the annual data are large and exhibit significant fluctuations, indicating a lack of consistency between years. - Consistent balance of payments series are not available for an adequate period of years. - When new source data, methodology, or statistical techniques have been introduced, historical revisions have not been made. - Major changes in methodology and unusual changes in economic trends have not been documented, and no analysis is included in regular balance of payments publications. - Balance of payments statistics are consistent with national accounts, government finance, and monetary statistics. ### Revision policy and practice - Revisions to balance of payments statistics do not have an established schedule and are not identified in the published statistics. - Explanations and analysis of the revisions are neither published nor available in internal documents of the SDS. - Preliminary data are not identified; revised data are disseminated with the same level of detail as originally published and without indication of which statistics have been updated. - External users are not informed of the causes of revisions. - Analysis of preliminary versus revised data is not published for major aggregates; direction and magnitude of revisions are not investigated; no record is kept on revisions. ### Data accessibility (presentation, dissemination, release) - A balance of payments publication including detailed data, charts, and tables was last published in 1997 in the format of BPM5. - Since 1997, quarterly and annual data are released without accompanying text, tables, and charts. - Dissemination media and formats do not meet users’ needs: there is no advance release calendar, no press release, and no mechanism for simultaneous release of data. - Only summary data are published. The first published release is through the NBG’s Quarterly Bulletin. - Summary data are published by the SDS in the Statistical Yearbook of Georgia or SDS Yearbook (the 2002 issue includes balance of payments data for 1997–2000). - In absence of a press release, finalized data (see Timeliness) are available upon request and routinely sent to NBG, MOF, the State Chancellery, the Ministry of Economy, Industry and Trade, and TACIS. - An unpublished full classification/sectorization, consistent with BPM5 recommendations, is sent to the IMF Statistics Department. - A schedule for data release is not announced in advance. - Balance of payments statistics are not made available to all users at the same time. - Nonpublished and nonconfidential sub-aggregates are provided upon request, usually for a fee; terms and conditions are not publicized. ### Metadata accessibility and user assistance - There is no internal documentation on concepts, scope, classifications, basis of recording, data sources, and statistical techniques used to compile balance of payments data. - Unlike most countries, a description of methods used for compiling the balance of payments is not published in the IMF’s Balance of Payments Statistics Yearbook. - Metadata are not adapted to the needs of the intended audience. - All statistical releases and publications indicate mailing address, telephone, fax, and e-mail of the SDS Publication Division and specific divisions (including BOPD); BOPD staff handle enquiries. - A catalogue of publications is available but does not include information on charges; information on how publications can be obtained is provided. ### Key deficiencies (from Data Quality Assessment Framework — Table 6) - Legal/institutional: Financial penalties for non or late response are nominal and not applied; potential confidentiality conflict between the Law on Statistics and the Anti-monopoly Law of Georgia; SDS lacks authority over regional offices; SDS does not build working relationships with respondents. - Resources: Computing and financial resources allocated are inadequate. - Quality awareness: No body charged with oversight or guidance on quality issues. - Transparency: Dissemination practices are not made known to the public; government agencies obtain access in advance; advance notice of methodology changes is not given. - Methodological soundness: Re-invested earnings are excluded; no instrument classifications for direct and portfolio investments; some data are on a cash basis. - Source data and statistical techniques: Data sources are inadequate; surveys are not conducted (except for an FDI survey with poor response); reliance on administrative sources; customs data deficient for the nonobserved economy; rule-of-thumb c.i.f. to f.o.b. adjustments; subjective estimates for shuttle imports; partner country data not used to adjust values; migration-related statistical model not regularly updated. - Validation and revision studies: Verification of relational data and trend analysis are not routinely performed; net errors and omissions are not analyzed; no incorporation of trade data reconciliation with partner countries; very little media monitoring; no revision records or studies. - Serviceability and accessibility: No established user consultation process; no advisory group; summary data released without text, charts, or comparable earlier years; no pre-announced schedule; priority access to government agencies ahead of general release. - Metadata accessibility: No information on sources and methods in SDS publications. - Assistance to users: Catalogue available but lacks charge information; publication access information provided. *Source: _cr03140 - 4.2 Timeliness and periodicity (compilation from the provided IMF document)._ ### 2. Methodological ### 2. Methodological ### Methodological soundness — concepts, scope, classification, basis for recording - The methodological basis for the statistics follows internationally accepted standards, guidelines, or good practices. - 2.1 Concepts and definitions – Concepts and definitions used are in accord with internationally accepted statistical frameworks. - 2.1.1 The overall structure in terms of concepts and definitions follows internationally accepted standards, guidelines, or good practices: see dataset-specific framework. - 2.2 Scope – The scope is in accord with internationally accepted standards, guidelines, or good practices. - 2.2.1 The scope is broadly consistent with internationally accepted standards, guidelines, or good practices: see dataset-specific framework. - 2.3 Classification/sectorization – Classification and sectorization systems are in accord with internationally accepted standards, guidelines, or good practices. - 2.3.1 Classification/sectorization systems used are broadly consistent with internationally accepted standards, guidelines, or good practices: see dataset-specific framework. - 2.4 Basis for recording – Flows and stocks are valued and recorded according to internationally accepted standards, guidelines, or good practices. - 2.4.1 Market prices are used to value flows and stocks. - 2.4.2 Recording is done on an accrual basis. - 2.4.3 Grossing/netting procedures are broadly consistent with internationally accepted standards, guidelines, or good practices. ### Accuracy and reliability — source data, techniques, validation, revisions - Quality objective: Source data and statistical techniques are sound and statistical outputs sufficiently portray reality. - 3.1 Source data - 3.1.1 Source data are collected from comprehensive data collection programs that take into account country-specific conditions. - 3.1.2 Source data reasonably approximate the definitions, scope, classifications, valuation, and time of recording required. - 3.1.3 Source data are timely. - 3.2 Statistical techniques - 3.2.1 Data compilation employs sound statistical techniques. - 3.2.2 Other statistical procedures (e.g., data adjustments and transformations, and statistical analysis) employ sound statistical techniques. - 3.3 Assessment and validation of source data - 3.3.1 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 planning. - 3.4 Assessment and validation of intermediate data and statistical outputs - 3.4.1 Main intermediate data are validated against other information where applicable. - 3.4.2 Statistical discrepancies in intermediate data are assessed and investigated. - 3.4.3 Statistical discrepancies and other potential indicators of problems in statistical outputs are investigated. - 3.5 Revision studies - 3.5.1 Studies and analyses of revisions are carried out routinely and used to inform statistical processes. ### Serviceability — relevance, timeliness, consistency, revision policy - Quality objective: Statistics are relevant, timely, consistent, and follow a predictable revisions policy. - 4.1 Relevance - 4.1.1 The relevance and practical utility of existing statistics in meeting users’ needs are monitored. - 4.2 Timeliness and periodicity - 4.2.1 Timeliness follows dissemination standards. - 4.2.2 Periodicity follows dissemination standards. - 4.3 Consistency - 4.3.1 Statistics are consistent within the dataset (e.g., accounting identities observed). - 4.3.2 Statistics are consistent or reconcilable over a reasonable period of time. - 4.3.3 Statistics are consistent or reconcilable with those obtained through other data sources and/or statistical frameworks. - 4.4 Revision policy and practice - 4.4.1 Revisions follow a regular, well-established and transparent schedule. - 4.4.2 Preliminary data are clearly identified. - 4.4.3 Studies and analyses of revisions are made public. ### Accessibility — data, metadata, and user assistance - Quality objective: Data and metadata are easily available and assistance to users is adequate. - 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). - 5.1.2 Dissemination media and formats are adequate. - 5.1.3 Statistics are released on the pre-announced schedule. - 5.1.4 Statistics are made available to all users at the same time. - 5.1.5 Nonpublished (but nonconfidential) sub-aggregates are made available upon request. - 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. - 5.2.2 Levels of detail are adapted to the needs of the intended audience. - 5.3 Assistance to users - 5.3.1 Contact person for each subject field is publicized. - 5.3.2 Catalogues of publications, documents, and other services, including information on any charges, are widely available. ### Users’ views — informal survey of Georgia data users (July 15-31, 2002 mission) - Survey logistics and respondents - Questionnaires sent out: 13 - Responses (excl. not applicable): 11 - Response rate: 85% - Respondent composition: government (3 responses), international, regional, and bilateral aid agencies (5 responses), commercial banks (2 responses), private sector entities (1 response). - Which macroeconomic statistics are used regularly (number of responses; total 11; percentages preserved) - National accounts: 8 — 73% - Price statistics: 9 — 82% - Fiscal statistics: 9 — 82% - Balance of payments: 9 — 82% - Monetary statistics: 9 — 82% - Coverage satisfactory? - Yes: 4 — 36% - No: 7 — 64% - Detail provided satisfactory? - Yes: 5 — 45% - No: 6 — 55% - Frequency satisfactory? - Yes: 9 — 82% - No: 2 — 18% - Timeliness satisfactory? - Yes: 8 — 73% - No: 3 — 27% - Accuracy satisfactory? - Yes: 1 — 9% - No: 10 — 91% - Statistics easily accessible? - Yes: 7 — 64% - No: 4 — 36% - Methodology well explained? - Yes: 6 — 55% - No: 5 — 45% - Enough information about revisions? - Yes: 4 — 36% - No: 7 — 64% - Overall assessment of Georgia’s statistics - Very good: 0 — 0% - Good: 3 — 27% - Unsatisfactory: 8 — 73% - Bad: 0 — 0% - Note: Where no opinion was offered to a particular question, the response was assumed to be favorable. ### Users’ comments and recurring concerns - Requests and suggested improvements - National accounts data for 1990-95 should be adjusted to be consistent with the data from 1996. - Some government users proposed a unified database for government use updated monthly, distinct from a less detailed public database; government users should have access to more detailed and more frequently updated data, possibly via direct data transmission. - Data quality concerns - Widespread concern about the reliability of data for components of GDP by expenditure, especially imports and exports of goods and services, and gross capital formation. - Concerns that economic and functional classification of government expenditure be consolidated to include special funds and local government. - Widespread recognition of the connection between accuracy and problems of poor response to statistical surveys and lack of information about the unobserved economy. - Meeting-specific points (government users and banks) - Both government users and banks considered that lack of information on the large unobserved economy has weakened the quality of national accounts and balance of payments data; some banks considered the unobserved economy much larger than official data indicate. - Both government users and banks considered that resource constraints in the SDS have seriously weakened the quality of national accounts and balance of payments data. - Both groups would benefit from access to complete macroeconomic databases that can be readily downloaded into excel spreadsheets, either on request or via official websites. - Banks criticized the format of statistical publications, which forces them to request additional information. - Widespread recognition that official data on unemployment are not meaningful. - Banks preferred access to monthly monetary data from the NBG’s website rather than quarterly statistical publications, and suggested that the NBG’s quarterly statistical publication provide more information on apparent inconsistencies between the tables. *Source: _cr03140 - 2. Methodological* --- _Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2003/_cr03140.pdf_