## _cr03256 — Executive Summary and Data Quality Assessment (Ukraine)

## Source details

**Canonical URL:** [_cr03256 — Executive Summary and Data Quality Assessment (Ukraine)](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2003/_cr03256.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2003/_cr03256.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2003/_cr03256.pdf.json)

---

### Executive Summary — Main conclusions
- Assessment scope:
  - ROSC—Data Module assesses Ukraine’s data dissemination practices against the IMF’s Special Data Dissemination Standard (SDDS) and provides in-depth assessment of data quality for national accounts, consumer and producer prices, government finance, monetary, and balance of payments statistics.
  - Agencies covered: State Statistics Committee (SSCU), Ministry of Finance (MoF), State Treasury of Ukraine (STU), National Bank of Ukraine (NBU).
  - Mission visit: April 3–17, 2002.
- SDDS status and key dissemination gaps:
  - Ukraine had not subscribed to SDDS at mission time but worked toward subscription; Ukraine subscribed in January 2003 (noted in integrity section).
  - Meets SDDS periodicity for all categories; meets or exceeds coverage for many categories.
  - Does not yet disseminate:
    - the Template on International Reserves and Foreign Currency Liquidity (Reserves Template), and
    - the international investment position (IIP).
  - NSDP and ARCs posted but not yet meeting SDDS requirements (not regularly updated).
- Integrity and institutional environment:
  - Legal/institutional framework supports statistical quality; laws guarantee professional independence.
  - Transparency generally good, but:
    - internal government access to monetary data ahead of public release not readily available publicly;
    - advance notice of major changes in GFS methodology not readily available to the public.
- Methodological frameworks:
  - National accounts: 1993 SNA.
  - Balance of payments: BPM5.
  - GFS: based on 1986 GFS Manual; migration planned to 2001 GFS Manual.
  - Monetary statistics: based on 1984 Guide; migration planned to MFSM 2000.
- Major coverage limitations:
  - Full scope of national accounts could not be implemented owing to source data constraints (residents working abroad insufficiently covered).
  - Domestic publications often exclude nonbudget pension and social security funds from GFS; these funds account for about one quarter of general government expenditure.
- Accuracy and reliability highlights:
  - Fourth quarter national accounts compiled using source data for the entire year and national accounts estimates for the first three quarters (residual fourth quarter practice).
  - Balance of payments primarily based on an ITRS for resident banks supplemented by enterprise direct reporting for transactions through nonresident banks.
  - Bank reporting threshold for payment orders was $250,000 (planned reduction to $50,000).

*Source: Executive Summary, ROSC—Data Module mission report for Ukraine (mission visit April 3–17, 2002).*

### SDDS observance — Key elements and timelines
- Data dimension:
  - Dissemination of 18 monitorable data categories (from 21 SDDS categories inclusive of external debt).
- Access dimension:
  - ARCs required: at least one-quarter ahead approximate dates and one-week ahead precise dates; simultaneous release to all users.
- Integrity dimension:
  - Required dissemination of terms and conditions, identification of internal government access, ministerial commentary identification, information about revisions and advance notice of methodology changes.
- Quality dimension:
  - Dissemination of documentation on methodology and component detail to permit cross-checks.
- Transition periods:
  - IIP transition period ends December 31, 2001; end-year 2001 IIP must be disseminated by June 30, 2002 (or Sept 30, 2002 under quarterly external debt option).
  - External debt transition ends March 31, 2003; quarterly external debt with one-quarter timeliness: quarter ending June 30, 2003 must be disseminated by end-September 2003.
- Selected dissemination practice exceptions (verbatim examples from table highlights):
  - National Accounts: SDDS Q Q Q 90–120 days — Data not in time-series format; Timeliness does not meet SDDS requirements; Ukraine has indicated intention to take a flexibility option for timeliness.
  - Consumer price index: SDDS M M M 10 days — Data not in time-series format; ARC shows range of dates but is not regularly updated.
  - General government operations: SDDS A A 2Q 6M — Social security funds not covered; does not meet SDDS.
  - International reserves template: Reserves template: NO — The data are not disseminated (reserves template).
  - International investment position: The data are not disseminated.

*Source: BOX 1 UKRAINE—DIMENSIONS AND ELEMENTS OF THE SPECIAL DATA DISSEMINATION STANDARD.*

### Prerequisites of quality — Institutional, staffing, resources
- Legal/institutional:
  - Statistics Law (July 13, 2000 No. 1922–III) defines roles; Presidential Decree April 14, 1995 (No. 312/95) establishes SSCU as central executive agency.
  - NBU Law No. 679–XIV (May 1999, amended Jan 10, 2002) provides legal basis for monetary statistics compilation.
- Coordination and data sharing deadlines (selected verbatim timing examples):
  - NBU monthly data: within 20 days after end of reporting month.
  - Preliminary annual balance of payments data: within 40 days after reporting year.
  - Quarterly balance of payments data: within 75 days after end of reporting quarter.
- Resources and organizational issues:
  - SSCU staffing: headquarters about 440 in Kyiv; regional network includes 25 oblast offices, 2 city offices, and 696 sub-offices.
  - Department of National Accounts divisions staff: Total staff 35 (university educated).
  - Price program staff budgeted positions: All price program positions: 570; Headquarters (Price Statistics Department): Total 21 (CPI 6; PPI 5; Construction Price Index 2; Retail/Communication/Transport indices 6; Administration 2); Computer Center: Total 9 (PPI 6; Construction 1; Retail/Communication/Transport 2); Regions: 540 (Centers 308; Districts 232).
  - STU headquarters and regions: some 14,500 staff; TSA extension and centralized ledger system by end-2002 estimated to require increase of about 4,500 staff and computers.
  - Monetary and Banking Statistics Division staffing: total 24 staff, including 11 professionals primarily devoted to monetary statistics.
  - Balance of Payments Department (BOPD) staff size: 24 employees.
- Computing and data processing:
  - SSCU: headquarters LAN Pentium PCs; regional computing barely adequate; PPI regional data entry uses custom Clipper applications.
  - NBU: highly computerized reporting; banks submit returns electronically with automatic edits; NBU final monthly balance sheet produced about 15 days after reference month due to internal organizations’ delays.

*Source: Various sections on prerequisites, resources, and 0.2 Resources.*

### Integrity — Professionalism, transparency, ethical standards
- Professionalism and independence:
  - Laws guarantee professional independence of staff; Article 5 Statistics Law and NBU Law provide autonomy.
  - Official users have exercised undue influence in certain cases (national accounts dissemination/revision practices).
- Transparency issues:
  - Advance notice of major changes: published in annual Budget Law but not in a readily accessible form for GFS users.
  - Monetary statistics are released to government officials about 10 days prior to public release; public awareness of this access is limited.
  - NSDP and ARCs posted but not regularly updated during mission.
- Ethical standards:
  - Ethics promoted via Civil Service Code, compilation laws, NBU Law Articles 65–66 and Law on Banks Articles 60–62.
  - Penalties for disclosure under administrative violations: fines for individuals 3–5 times the pre-tax minimum personal income; for legal units 10–15 times the pre-tax minimum personal income (verbatim from multiple sections).

*Source: Integrity sections and legal references.*

### Methodological soundness — Key technical observations
- National accounts:
  - Framework: 1993 SNA; classifications largely follow international recommendations.
  - Fourth quarter national accounts previously compiled residually from entire year source data and first three quarters estimates; work underway to compile Q4 from source data consistent with first three quarters.
  - Volume measures and index series not disseminated in time-series format tied to fixed reference period; SSCU plans recalculation to fixed base (2001 =100).
  - Lack of quarterly export/import price indices limits expenditure-side volume measures.
- Price indices:
  - CPI concepts consistent with 1993 SNA and ILO Manual (1989); CPI covers urban population about 68 percent of total; no owner-occupied housing coverage.
  - PPI covers 78 percent of output of covered industries; inclusion of more market services desirable.
  - Index formula: Laspeyres at all aggregation levels; imputation of missing prices done only one way—by product.
  - Release timeliness: CPI and PPI released within 10 days of end of reference month.
- Government finance statistics:
  - Based on GFSM 1986; migration plan to GFSM 2001 in place.
  - Published monthly time series relate to budgetary central government; quarterly statistics relate to general government.
  - Coverage shortfall: nonbudget pension and social security funds excluded from domestic GFS publications—about one quarter of general government expenditure.
- Monetary statistics:
  - Framework based on IMF draft Guide (1984); MFSM 2000 adoption planned.
  - Depository corporations include NBU and 154 active commercial banks (as of Dec 31, 2001); monetary aggregates M0, M1, M2, M3 defined verbatim.
  - Exclusion of 35 commercial banks in liquidation from monetary aggregates; memorandum items provided.
- Balance of payments:
  - Compiled in broad conformity with BPM5; sectorization and classifications consistent with BPM5.
  - ITRS closed-bank reporting system complemented by enterprise direct reporting and SSCU surveys.
  - Shuttle trade adjustments: imports adjusted via data model; exports via bilateral exchanges; quality would be improved by travelers’ survey.
  - Payment order reporting threshold: banks required to provide copy of payment order for transactions over US$250,000 (mission); NBU planned reduction to US$50,000 (noted as intended and later board resolution No 124 of 19.03.2003).

*Source: Methodological soundness sections across datasets.*

### Accuracy and reliability — Source data practices and validation
- National accounts:
  - Source data from SSCU survey program supplemented with administrative sources; 1999 surveys covered 88.9 percent of total output; in 2001 medium/large enterprises covered increased to 47,489.
  - Quarterly data not seasonally adjusted; revisions follow a regular schedule but revision procedures not fully explained publicly.
- CPI/PPI:
  - CPI weights from HLCS; HLCS processing lag eight months; CPI sample reduced from about 475,000 to 275,000 price records monthly.
  - PPI monthly industry survey: 50 percent cutoff sample by four-digit NACE plus judgmental sample; covers 80 percent of aggregate output of covered industries.
  - Imputation for missing prices currently not “self-correcting”; CPI published data are final and not revised.
- GFS:
  - Source: STU accounting records, State Debt Department debt database; Treasury Single Account (TSA) introduction phased (central government TSA complete July 1, 2002; local government TSA expected January 1, 2004).
  - Revisions minuscule; arrears data collected since 1998.
- Monetary statistics:
  - Source: NBU and commercial banks accounting records; fully computerized reporting; internal cross-checks; no routine seasonal adjustment.
  - Data final when first published; revision studies not routine or publicly disseminated.
- Balance of payments:
  - ITRS provides timely bank reporting; direct enterprise reporting via form 2–PB (~100 enterprises); inter-enterprise receivables/payables survey covers ~55,000 enterprises.
  - Identified source-data issues requiring action (verbatim priorities):
    - (i) Lower the threshold for banks to provide payment order information and ensure payment orders identify the resident party to the transaction.
    - (ii) Regularly conduct a travelers’ survey.
    - (iii) Revise inter-enterprise receivables/payables survey forms to include debt instruments, currency composition, debt maturity, write-offs.
    - (iv) Separately identify FDI investor loans in FDI survey forms.

*Source: Accuracy and reliability sections per dataset.*

### Serviceability and accessibility — Timeliness, periodicity, revisions, dissemination
- Timeliness/periodicity highlights (verbatim timings where given):
  - Quarterly BOP: generally disseminated with a lag of 75–80 days.
  - CPI and PPI: released within 10 days of end of reference month.
  - Timeliness of quarterly national accounts first three quarters: 95 days; fourth quarter disseminated around four months after end of reference quarter (mission: needs improvement to meet SDDS 90–120 days table note).
  - Monthly central government operations available ~25 days after reference period but released to public after 35 days.
  - Central government debt quarterly: published within 45 days.
  - NBU final monthly balance sheet produced about 15 days after the reference month; analytical accounts disseminated with lag ~30 days.
- Accessibility and metadata:
  - SSCU posted ARCs and NSDP but not regularly updated during mission; SSCU planned recalculation to fixed reference base (2001 =100) for volume and index series.
  - Data and metadata generally available in hard copy; many series available electronically on agency websites (SSCU, MoF, NBU).
  - Metadata in SDDS format not posted on MoF website at mission time.
  - Nonpublished nonconfidential data available upon request from contact persons; contact details published.
- Consistency and revision policies:
  - Quarterly BOP and national accounts revision timing differ (BOP revisions at 15 months; national accounts final revisions ~12 months), creating inconsistency; recommendation to harmonize revision schedules.
  - Quarterly data revised at fourth-quarter publication; BOPD revises current and previous year at that time.
  - Revision studies routinely conducted internally for some datasets but not regularly published with explanations.

*Source: Serviceability and Accessibility sections.*

### Agencies, dissemination channels, and contact points
- Four agencies compile macrostatistics: SSCU, MoF, STU, NBU.
- Online access (verbatim URLs as provided):
  - SSCU website: http://www.ukrstat.gov.ua (in Ukrainian)
  - MoF website: http://www.minfin.gov.ua (in Ukrainian)
  - NBU website: http://www.bank.gov.ua/ENGL/DEFAULT/htm (in English)
  - STU: does not provide data via a website.
  - Share price index compiled by First Securities Trading System: http://www.pfts.com/ukr

*Source: Agencies and dissemination channels section.*

### Staff mission and responses to recommendations (selected actions and dates)
- Mission team: Messrs. Di Calogero (head), Zieschang, Khawaja, Kucera, Piché, Doggett (expert), and Ms. Tayas (staff assistant).
- Authorities’ inputs (March 2002):
  - partial draft metadata for 20 of 21 prescribed data categories,
  - ARCs for 16 data categories,
  - summary methodology statements for 13 data categories.
- Authorities’ planned/implemented responses (selected verbatim commitments):
  - NSDP and ARCs: NSDP posted and regularly updated on SSCU website; ARC information updated regularly.
  - Reserves Template: compiled in line with SDDS and disseminated on NBU website.
  - Interagency committee for quarterly external debt data: NBU and MoF responsibility to be fixed after government approves relevant legislation acts.
  - Volume measures and index series: recalculated to fixed reference period (2001 =100).
  - Users’ surveys: planned as part of “Ukraine Statistics. Strategy and Implementation Plan. 2004-2008”.
  - TSA and STU: Treasury local budgets servicing started on January 1, 2002; stage-by-stage introduction planned up to 2005.
  - NBU board resolution No 124 of 19.03.2003: changes in charts of accounts to be made once a quarter; also approval to reduce ITRS threshold from 250 to 50 thousands U.S. dollars (NBU intended in 2003 to reduce reporting threshold).

*Source: II. Response to Staff’s Recommendations and dataset-specific responses.*

### Staff recommendations — General and dataset-specific (verbatim/high-priority items)
- General (selected):
  - Regularly update information in NSDP and ARCs on SSCU website.
  - Disseminate the Data Template on International Reserves and Foreign Currency Liquidity and the IIP.
  - Establish interagency committee for quarterly external debt data assignment.
  - Disseminate volume measures and index series in relation to a fixed reference base.
  - Conduct regular users’ surveys; provide advance notice of major methodology/source changes.
- National Accounts — High priority:
  - Greater use of sample surveys and improved data flow management and processing.
  - Revive the high-powered Statistical Council to guide quality/improvement strategies.
  - Strengthen SSCU independence to prevent undue official influence on methodology, dissemination, revision policies.
  - Compile seasonally adjusted measures of GDP.
  - Compile fourth quarter national accounts from source data consistent with first three quarters; improve Q4 timeliness to meet SDDS (Q4 to be compiled on the 90th day after reference quarter per response).
- CPI — High priority:
  - Implement procedures in regional offices to impute prices of each missing variety month-to-month until permanently missing; particularly for seasonal items (fresh fruits and vegetables).
- GFS — High priority:
  - Re-disseminate annual general government data for 1999–2001 in IMF GFS Yearbook format including nonbudget pension/social security funds.
  - Publish more extensive documentation of definitions, sources, methods.
  - Incorporate nonbudget social security funds in disseminated central government operations data and improve timeliness.
- Monetary Statistics — High priority:
  - Improve timeliness of analytical accounts of central bank to meet SDDS using daily accounting records excluding internal organizations.
  - Disseminate information on compilation of monetary data including differences from international standards.
  - Establish migration plan to adopt MFSM 2000 recommendations.
- Balance of Payments — High priority:
  - Implement a lower reporting threshold for banks to provide payment order information for high-value transactions (from $250,000 to $50,000 planned).
  - Implement a travelers’ survey to improve shuttle trade, compensation of employees, and travel estimates.
  - Expand inter-enterprise arrears survey form to include financial instruments, currency composition, debt maturity, write-offs; consider quarterly frequency and sampling techniques.
  - Prepare and regularly publish revision studies explaining sources and magnitude of revisions.

*Source: Staff’s recommendations sections and dataset-specific recommendations.*

### Users’ survey — Key findings and numeric summaries (verbatim figures)
- Survey respondents: TOTAL 30 (Government 13; Universities/Research Institutes 11; Banks 1; Private Sector 5).
- Uses of official macroeconomic statistics (number of respondents indicating regular use):
  - National accounts: 21
  - Production indices: 23
  - Prices: 22
  - Labor market: 22
  - Balance of payments: 19
  - Merchandise trade: 22
  - International reserves and foreign liquidity: 13
  - External debt: 17
  - International investment position: 12
  - Government/public sector operations and debt: 16
  - Monetary and financial statistics: 22
- Satisfaction indicators (selected):
  - Know public release calendar? YES: 13; NO: 17.
  - Can easily access official statistics? YES: 19; NO: 11.
  - Is methodology information sufficiently clear and detailed? YES: 14; NO: 11; NO OPINION: 4.
  - Is there enough information about revisions? YES: 6; NO: 15; NO OPINION: 6.
- Main user concerns (verbatim highlights):
  - Nonavailability of data in time series format (fixed reference period) for GDP, CPI, PPI, industrial production index.
  - Publication of data on cumulative basis (e.g., monthly merchandise trade) is not appropriate.
  - Requests for more detailed SNA components, improved GDP deflation methodology, reconciliations between datasets, better revision explanations.
  - Electronic data availability upon special request satisfactory but SSCU website lacks up-to-date data retrieval; NBU website cited as best statistical website of Ukraine.

*Source: Users’ Survey section.*

### DQAF summary — Overall assessment highlights (verbatim summary indicators)
- Overall conclusion: Ukraine’s macroeconomic statistics and statistical base are adequate to conduct effective surveillance, but shortcomings exist that could detract from accurate and timely analysis and policy formulation.
- DQAF summary indicators (selected element summaries):
  - Source data: available provide an adequate basis to compile statistics.
  - Statistical techniques: conform to sound statistical procedures.
  - Assessment and validation: source data and intermediate data are routinely assessed and validated, but routine public revision studies often lacking.
  - Serviceability: data generally relevant and timely though specific timeliness exceptions and consistency issues noted (e.g., Q4 national accounts lag; BOP vs NA revision timing).
  - Accessibility: data and metadata generally available, but time-series in fixed base not consistently disseminated; ARCs not regularly updated during mission.

*Source: DQAF-based summary and sections on 3. Accuracy and reliability; 4. Serviceability; 5. Accessibility.*

---

*Source: IMF staff report (excerpts), ROSC—Data Module mission report for Ukraine (mission visit April 3–17, 2002); BOX 1; DQAF assessments; dataset-specific sections as provided in content unit _cr03256.*

### Executive Summary.......................................................................................................

### _cr03256 - Executive Summary

### Executive Summary — Main conclusions
- This ROSC—Data Module assesses Ukraine’s data dissemination practices against the IMF’s Special Data Dissemination Standard (SDDS) and provides an in-depth assessment of data quality for national accounts, consumer and producer prices, government finance, monetary, and balance of payments statistics. The assessment covers statistics compiled by the State Statistics Committee (SSCU), the Ministry of Finance (MoF), the State Treasury of Ukraine (STU), and the National Bank of Ukraine (NBU).  
- The assessment was carried out by a mission from the IMF Statistics Department that visited Kyiv during April 3–17, 2002.

### SDDS observance
- Ukraine does not yet subscribe to the SDDS at the time of the mission but has been working closely with IMF staff toward subscription.
- Ukraine meets SDDS requirements for coverage of many data categories.
- Ukraine meets or exceeds SDDS periodicity for all categories.
- Ukraine meets timeliness requirements for most data categories, but does not yet disseminate:
  - the Template on International Reserves and Foreign Currency Liquidity, and
  - the international investment position (IIP).
- Ukraine has posted a National Summary Data Page (NSDP) and advance release calendars (ARCs) on the Internet, but these do not yet meet SDDS requirements.

### Prerequisites of quality
- Legal and institutional framework supports statistical quality; agencies demonstrate awareness of quality.
- Coordination among statistical agencies has a basis in law and in protocols/agreements on data sharing and provision of source data.
- Human and computer resources are adequate for present data series, but additional resources will be needed to implement development plans to meet international standards.
- The SSCU has excessive staff at regional levels that could be more effectively used at headquarters.

### Integrity
- Agencies demonstrate professionalism and staff are qualified.
- Laws guarantee professional independence of staff.
- Ethical standards are promoted via the Civil Service Code and compilation laws.
- Transparency is generally good, but:
  - information on internal government access to monetary data ahead of public release is not readily available, and
  - advance notice of major changes in government finance statistics (GFS) methodology is not readily available to the public.

### Methodological soundness
- National accounts follow System of National Accounts 1993 (1993 SNA).
- Balance of payments compiled according to Balance of Payments Manual, 5th edition (BPM5).
- GFS are based on the 1986 GFS Manual; plans are underway to migrate to the 2001 GFS Manual.
- Monetary statistics are based on the 1984 Guide; plans are underway to migrate to the 2000 Monetary and Financial Statistics Manual.
- Full scope of national accounts coverage cannot be implemented owing to source data constraints.
- Domestic publications often exclude nonbudget pension and social security funds from GFS; these funds account for about one quarter of general government expenditure.

### Accuracy and reliability
- Statistics are reliable and accurate with extensive source-data checks in most cases.
- Fourth quarter national accounts are compiled using source data for the entire year and national accounts estimates for the first three quarters.
- Revision studies are regularly conducted, but insufficient information on revision procedures is provided to the public.
- Balance of payments is based primarily on an International Transactions Reporting System (ITRS) for resident banks supplemented by direct reporting of enterprises for transactions through nonresident banks.
- Checks on accuracy of reporting could be enhanced by lowering the reporting threshold for banks to provide payment order information to the NBU.

### Serviceability
- Statistics are overall timely and relevant; periodicity meets or sometimes exceeds international standards (SDDS).
- Limitations:
  - Lack of consistent coverage of budget transactions from year to year.
  - Volume measures of GDP and its components are not disseminated in time series format—expressed in terms of a fixed reference period—limiting usability of data.
- Statistical agencies have not formally established procedures to regularly ascertain user needs.

### Accessibility
- SSCU has posted ARCs on the Internet for all SDDS data categories, but ARCs do not yet meet SDDS requirements as they are not regularly updated.
- Data and metadata are generally available in hard copy publications.
- Most data series are disseminated in electronic form on agency websites.
- Contact persons in the various agencies are identified.

### Mission and staff notes
- The mission team included Messrs. Di Calogero (head), Zieschang, Khawaja, Kucera, Piché, Doggett (expert), and Ms. Tayas (staff assistant).
- In March 2002, authorities provided:
  - partial draft metadata for 20 of 21 prescribed data categories,
  - advance release calendars (ARCs) for 16 data categories, and
  - summary methodology statements for 13 data categories.
- Some metadata submitted described dissemination practices planned for SDDS subscription but not yet implemented.

### Current dissemination practices and outstanding issues
- Authorities planned to subscribe to the SDDS during the second half of 2002.
- The NSDP posted by SSCU did not yet contain data for all SDDS categories; ARCs were not regularly updated and therefore did not fully meet SDDS requirements.
- Authorities were preparing the Data Template on International Reserves and Foreign Currency Liquidity and IIP data, but had not disseminated this information to the public at the time of the mission.
- Table 1 (in the source) compares Ukraine’s practices to SDDS requirements and identifies exceptions where coverage did not meet SDDS (national accounts, industrial production index, consumer and producer prices, merchandise trade, and general government operations due to incomplete coverage of nonbudget pension and social security funds).
- All data categories met or exceeded prescribed periodicity; exceptions to timeliness included national accounts, analytical accounts of the central bank, and central government operations.

### Agencies and dissemination channels
- Four agencies compile macroeconomic statistics: SSCU, MoF, STU, and NBU.
- Access to data online:
  - SSCU website: http://www.ukrstat.gov.ua (in Ukrainian)
  - MoF website: http://www.minfin.gov.ua (in Ukrainian)
  - NBU website: http://www.bank.gov.ua/ENGL/DEFAULT/htm (in English)
  - STU does not provide data via a website.
- The share price index is compiled and disseminated by the First Securities Trading System at http://www.pfts.com/ukr.

*Source: Executive Summary, ROSC—Data Module mission report for Ukraine (mission visit April 3–17, 2002).*

### BOX 1 UKRAINE—DIMENSIONS AND ELEMENTS OF THE SPECIAL DATA DISSEMINATION

### BOX 1 UKRAINE—DIMENSIONS AND ELEMENTS OF THE SPECIAL DATA DISSEMINATION STANDARD

### Data dimension (coverage, periodicity, and timeliness)
- Dissemination of 18 monitorable data categories (from 21 SDDS data categories inclusive of external debt data), including component detail, covering the four main macroeconomic statistical sectors, with prescribed periodicity and timeliness.

### Access dimension
- Dissemination of advance release calendars providing at least a one-quarter ahead notice of approximate release dates, and at least a one-week ahead notice of the precise release dates.
- Simultaneous release of data to all users.

### Integrity dimension (dissemination of institutional information)
- Dissemination of the terms and conditions under which official statistics are produced and disseminated.
- Identification of internal government access to data before release.
- Identification of ministerial commentary on the occasion of statistical release.
- Provision of information about revision and advance notice of major changes in methodology.

### Quality dimension
- Dissemination of documentation on statistical methodology and sources used in preparing statistics.
- Dissemination of component detail and/or additional data series that make possible cross-checks and checks of reasonableness.

### Subscriber requirements and transition periods
- Subscribers are required to post description of their data dissemination practices (metadata) on the IMF’s Dissemination Standards Bulletin Board (DSBB). Summary methodology statements are disseminated on the DSBB.
- Subscribers must maintain an Internet website, referred to as the National Summary Data Page, which contains the actual data described in the metadata, and to which the DSBB is electronically linked.
- Transition period for the International Investment Position (IIP):
  - Transition period ends on December 31, 2001.
  - IIP data are to be disseminated with annual periodicity and six months timeliness; data for the end-year 2001 must be disseminated no later than June 30, 2002 (or September 30, 2002 provided the subscriber is disseminating quarterly external debt data with a one-quarter lag).
  - If IIP data are based on a fiscal year not equal to the calendar year, the first reference period is the fiscal year ending after June 30, 2001; dissemination deadline is six months (or nine months) after the end of the reference period (example: fiscal year ending September 30, 2001 → dissemination deadline March 31, 2002 or June 30, 2002 if external debt data are disseminated quarterly with a one-quarter lag).
- Transition period for the external debt data category:
  - Transition period ends on March 31, 2003.
  - For quarterly external debt data with one-quarter timeliness, data for the quarter ending June 30, 2003 must be disseminated by end-September 2003.

### Selected data dissemination practices (table highlights)
- Real Sector (selected entries)
  - National Accounts: SDDS Q Q Q 90–120 days — Data not in time-series format; Timeliness does not meet SDDS requirements; Ukraine has indicated intention to take a flexibility option for timeliness.
  - Production Index: SDDS M M M 6W (M encouraged) — Data not in time-series format; 2 weeks ARC shows range of dates but is not regularly updated.
  - Employment: SDDS Q Q Q 60 days — Meets SDDS.
  - Unemployment: SDDS Q Q Q 60 days — Meets SDDS.
  - Wages/earnings: SDDS Q M Q 26 days — Meets SDDS.
  - Consumer price index: SDDS M M M 10 days — Data not in time-series format; ARC shows range of dates but is not regularly updated.
  - Producer price index: SDDS M M M 10 days — Data not in time-series format; ARC shows range of dates but is not regularly updated.
- Fiscal Sector (selected entries)
  - General government operations: SDDS A A 2Q 6M — Social security funds not covered; does not meet SDDS.
  - Central government operations: SDDS M M M 2Q 35 days — Meets SDDS coverage but timeliness does not meet SDDS requirements; existing practices not yet aligned with the ARC.
  - Central government debt: SDDS Q Q Q 45 days — Meets SDDS; ARC shows no-later-than dates but is not regularly updated.
- Financial Sector (selected entries)
  - Analytical accounts of the banking sector: SDDS M M M M M for the first 11 months. Within 6 weeks for December data. ARC shows no-later-than dates but is not regularly updated.
  - Analytical accounts of the central bank: SDDS M (W recommended) M 2W (W encouraged) M — Timeliness does not meet SDDS requirements; Ukraine availed itself of a flexibility option for timeliness of central bank analytical accounts at subscription.
  - Interest rates: SDDS D D ... D — Meets SDDS; ARC not applicable.
  - Stock market: SDDS D D ... D — Meets SDDS; ARC not applicable.
- External Sector (selected entries)
  - Balance of payments: SDDS Q Q Q 75–80 days — Meets SDDS.
  - International reserves and foreign currency liquidity: Official reserves: SDDS M (W recommended) M NA W M NA — Reserves template: NO — ARC shows no-later-than dates but is not regularly updated; The data are not disseminated (reserves template).
  - Merchandise trade: SDDS M M M 8W (4–6W encouraged) 45 days — Cumulative data; ARC shows a range of dates but is not regularly updated.
  - International investment position: SDDS A (Q recommended) NA 2Q (Q encouraged) NA — The data are not disseminated.
  - Exchange rates: SDDS D D ... D — Meets SDDS; ARC not applicable.
  - External debt: SDDS Q NA Q NA — Transition period ends on March 31, 2003.
  - Addendum — Population: SDDS A A NA 45 days — ARC not applicable.

### Summary assessment of data quality (DQAF-based)
- Overall conclusion
  - Ukraine’s macroeconomic statistics and statistical base are adequate to conduct effective surveillance, but shortcomings exist that could detract from accurate and timely analysis and policy formulation.

- Prerequisites of quality (institutional aspects and resources)
  - SSCU responsibilities: SSCU compiles and disseminates national accounts, CPI, PPI, and other statistics; role set out in the Statistics Law.
  - Legal/institutional provisions: Decrees/resolutions/orders of the President and Cabinet of Ministers specify responsibilities and ensure coordination among data-producing agencies. Confidentiality of respondents’ data is protected in the Statistics Law.
  - Resource and organizational issues:
    - Staff and technological resources need supplementation to implement development programs (e.g., compiling financial accounts and comprehensive supply-use tables).
    - SSCU has excessive staff in the regions and overreliance on complete census or very large counts; data processing involves duplication; reallocation to headquarters suggested.
    - No body distinct from the SSCU currently provides guidance on statistical series quality and improvement strategies; revival of the high powered Statistical Council previously attached to the office of the President could be beneficial.
  - Fiscal statistics responsibilities: Formally allocated among MoF, STU, and SSCU, with MoF lead; specified in the Budget Code (enacted June 21, 2001), Order of the MoF No. 265 amended, and the Statistics Law.
    - Current MoF staff sufficient for present needs; more resources needed to migrate to the 2001 GFS Manual and meet SDDS requirements.
    - Treasury Single Account (TSA) introduction expected complete for central government from July 1, 2002; local government transactions should be fully covered from January 1, 2004.
    - STU headquarters and regions: some 14,500 staff; extension of TSA and planned centralized ledger accounting system by end-2002 estimated to require an increase of about 4,500 staff and computers.
    - Concern about ability of STU to meet implementation dates given inadequate and aging computer stock; an efficiency review of the STU would be necessary after TSA extension.
  - Monetary and balance of payments statistics: Responsibility of the NBU under Law of Ukraine on the National Bank of Ukraine, No. 679-XIV of May 1999 (amended January 10, 2002).
    - Individual reporting data confidentiality established; Economic Department of the NBU responsible for methodology, compilation, processing, dissemination.
    - Staffing and qualifications are adequate for existing tasks; turnover not high; sufficient personal computers for existing tasks.
    - Resources insufficient to expand institutional coverage of monetary accounts and compile the flow of funds for the financial sector.
    - Managers and staff are sensitive to data quality; improvement is a focus in planning.

- Integrity
  - National accounts and price statistics: Statistics Law provides legislative basis for professional independence and forbids interference, but official users have continued to exercise undue influence over dissemination practices or revision policies; price publications give no advance notice of weight updates and methodology changes.
  - Government finance statistics: MoF and STU staff have appropriate professional independence; no government ministers or officials have access to data before release. MoF publications identify individuals involved. Advance notice of major methodology changes is published in the annual Budget Law but not in readily accessible form.
  - Monetary and balance of payments statistics: Compiled impartially; guidelines for ethical standards in the Law on the NBU and the Law on Banks and Banking Activity. Major methodological changes are announced in advance for monetary statistics and at the time of introduction for balance of payments. Data are confidential until issuance of the monthly Bulletin of the NBU and dissemination on the NBU website; however, monetary statistics are released to government officials for internal analysis about 10 days prior to public release and there is no public awareness of this.

- Methodological soundness
  - National accounts: 1993 SNA is the general framework; classifications largely follow internationally recommended systems. Delimitation of economy/production/assets observed in principle but not fully implemented due to source data constraints (e.g., lack of adequate coverage of residents working abroad).
  - Price indices:
    - CPI concepts and definitions fit within 1993 SNA and Consumer Price Indices; An ILO Manual (1989); index covers urban population (about 68 percent of total); no coverage of owner occupied housing.
    - PPI concepts and definitions fit within 1993 SNA and broadly follow good practice; PPI covers 78 percent of the output of the covered industries; industry coverage conventional (mining, manufacturing, utilities, energy production); inclusion of more market services desirable.
    - Both indices implement international classification standards for goods and services, expenditures, institutional units, and establishments.
  - Fiscal statistics: Generally follow 1986 GFS Manual concepts and definitions; plans to migrate to 2001 GFS Manual drawn up.
    - From 2002 revised Classification of Functions of Government to be used for central government but not yet extended to local government.
    - Privatization receipts to be treated as financing.
    - Published monthly time series relate to budgetary central government; quarterly statistics relate to general government.
    - Coverage of the budget has varied and is significantly less than GFS Manual coverage in all years. Main deficiency: exclusion of nonbudget pension and social security funds—about one quarter of general government expenditure—from data published domestically.
  - Monetary statistics: Analytical framework generally based on IMF’s draft Guide to Money and Banking Statistics in International Financial Statistics (December 1984).
    - Data do not cover banks in the process of liquidation.
    - Following MFSM publication (September 2000), NBU intends to revise procedures and formats to comply with the Monetary and Financial Statistics Manual (MFSM), including:
      - Expanding coverage to include accounts of other financial corporations that issue deposit substitutes.
      - Separate identification of financial derivatives on-balance sheet.
      - Adoption of terminology consistent with 1993 SNA and MFSM.
      - Compilation of all components of the flow data.

*Source: BOX 1 UKRAINE—DIMENSIONS AND ELEMENTS OF THE SPECIAL DATA DISSEMINATION STANDARD*

### 22.      Ukraine’s balance of payments statistics are compiled in broad conformity with the

### 22. Ukraine’s balance of payments statistics are compiled in broad conformity with the

### Coverage and compilation framework
- Balance of payments statistics are compiled in broad conformity with the conceptual framework of the BPM5.
- Sectorization of institutional units and classification of transactions are consistent with the BPM5.
- Component details for current transfers are not provided.
- Inter-enterprises’ receivables and payables are not broken down between trade credits, direct investment loans, and other.
- Transactions recorded at market prices and converted to U.S. dollars using average monthly rates for ITRS data.
- Data compilation is, in principle, on an accrual basis.
- Shuttle trade adjustments: imports adjusted using a data model; exports adjusted via bilateral exchange of information; quality would be improved using a travelers’ survey.
- Coverage of residents working abroad is insufficient.

### Accuracy and reliability — source data and methods
- National accounts
  - Source data come from the extensive survey program of the SSCU, supplemented with administrative and other sources.
  - Main survey report forms revised recently to align with national accounts definitions and the new chart of accounts.
  - Fourth quarter national accounts compiled using source data for the whole year and national accounts data of the first three quarters.
  - Quarterly data are not seasonally adjusted.
  - Revision studies are conducted, but revision procedures are not explained.
  - Volume index for manufacturing compiled residually from enterprise-reported constant-price data.
  - No proper quarterly price indices for exports and imports.
- CPI and PPI
  - Source data follow standard stratified sample design with cutoff and purposive selections.
  - Indexes use standard Laspeyres index methodologies.
  - Weights for CPI and PPI updated annually with previous year data; linking: PPI in July and CPI in August.
  - No formal routine assessments of sample or nonsampling error; follow-up on nonrespondents is practiced.
  - Procedure for handling missing prices for seasonal/temporarily unavailable goods needs improvement.
- Government Finance Statistics (GFS)
  - Budgetary system operates on a cash basis; source data aggregated from accounting records.
  - Since 1998, data on arrears of revenue and expenditure collected, classified by economic category and functions.
  - Data on outstanding debt taken from the debt database of the State Debt Department.
  - Revisions are minuscule.
- Monetary statistics
  - Source data derived from accounting records of the NBU and commercial banks; cover full range of financial instruments and sectors.
  - NBU does not calculate seasonally adjusted monetary data but is examining EUROSTAT methods.
  - Internal arrangements ensure information flow and quality through cross-checks.
  - Data are final when first published; revision studies are not relevant.
- Balance of payments specific sources and issues
  - NBU data collection based on banks’ reporting for transactions through resident banks and direct enterprise reporting for transactions through nonresident banks; supplemented with SSCU survey information.
  - High-value transactions are monitored systematically, but bank reporting threshold for payment order information is $250,000 and does not allow monitoring classification accuracy below that level.
  - NBU expected to shortly implement a $50,000 threshold.
  - Merchandise trade data based on customs reporting.
  - Survey on stock of inter-enterprises’ receivables and payables used to derive selected other investment flows, but absence of survey info on debt instruments, currency composition, debt maturity, and write-offs reduces quality.
  - NBU investigates sources of errors and revisions, but studies/explanations on methods of revising data are not made available.

### Serviceability — timeliness, periodicity, consistency, revisions
- National accounts
  - User feedback obtained officially via working groups and informally; users’ surveys not conducted regularly.
  - Fourth quarter data disseminated more than three months after end of reference quarter; need greater timeliness to meet SDDS requirements.
  - Periodicity meets SDDS requirements.
  - Revisions follow a regular schedule, but one-year revision period does not permit incorporation of balance of payments statistics finalized after 15 months.
  - Quarterly national accounts unable to use household survey results prepared four months after end of reference quarter.
  - Statistical discrepancy often less than 2 percent of GDP; not shown explicitly and removed through adjustments in weaker data-source components.
  - National accounts for 2000–2001 compiled using new national classification; no retrospective adjustments for previous years.
  - Deflation methodology for quarterly GDP not consistent with dissemination of volume measures expressed in terms of a fixed reference period.
- Price statistics
  - Meet SDDS requirements for periodicity and timeliness.
  - CPI treatment of seasonally available items could cause drift, especially for fresh fruits and vegetables.
  - No formal advisory group or regular users’ survey; no published study of impact of annual CPI-PPI weight updates.
- GFS
  - Usefulness diminished by changes in coverage year-to-year; trend toward increasing coverage.
  - Periodicity and timeliness meet SDDS requirements, except central government operations available with a lag of 25 days but only released after 35 days.
  - No formal surveys of user requirements.
- Monetary statistics
  - Analytical accounts disseminated with a lag of about 30 days after end of reference month.
  - NBU could meet SDDS timeliness for central bank analytical accounts: preliminary aggregates (foreign positions, reserve money, and domestic credit) available within two weeks after reference period.
  - Consistency checks with balance of payments done regularly; not undertaken with government finance statistics.
  - Frequent changes in charts of accounts of NBU and commercial banks adversely impact consistency over time.
- Balance of payments
  - Major aggregates consistent over time; introduction of new data sources/methodologies leads to backward revisions to previous two years per NBU revision policy.
  - Net errors and omissions significant during 1997-99 but decreased during last two years.
  - Balance of payments and national accounts data sometimes inconsistent due to different revision timing.
  - Data for the current and previous year considered preliminary and subject to revision; policy in place but not publicly described.
  - NBU does not conduct users’ surveys to assess relevance of statistics.
  - Significant revisions are footnoted in tables, but additional information on adjustments and revisions is needed.
  - Quarterly BOP data generally disseminated with a lag of 75–80 days.

### Accessibility — dissemination and metadata
- National accounts, CPI, PPI dissemination and metadata not fully adequate: not disseminated in time-series format tied to a fixed reference period.
- National accounts data and metadata available in publications and electronic form; CPI and PPI documentation available but not timely or comprehensive.
- Nonpublished nonconfidential data available upon request from contact persons.
- SSCU developing an advance release calendar including fiscal data release dates; not yet operational.
- Monthly fiscal statistics format aimed at short-term budget monitoring; assembling data for broader fiscal analysis requires substantial effort; long-term analysis virtually impossible.
- Documentation exists mainly in Budget Code and Laws and summary notes; no comprehensive methodology presentation.
- Metadata in SDDS format not posted on MoF website.
- Monetary statistics disseminated in hard copy and electronic format; more detailed data in monthly Bulletin of NBU.
- No ARC for monetary statistics; NBU follows internal publication schedule known to users.
- Unpublished nonconfidential data available upon request free of charge.
- Monthly Bulletin contains short description of terms but not the full compilation framework.
- NBU’s Balance of Payments of Ukraine publication: quarterly BOP data per BPM5, additional tables on merchandise trade and shuttle trade, analysis of trends, and notes on sources and methods.
- Few additional BOP details published; SSCU can provide direct investment and merchandise trade details though they may be inconsistent with BOP due to NBU adjustments to comply with BPM5.

### Staff’s recommendations — General
- Take remaining measures to be able to subscribe to the SDDS, including:
  - Regularly updating information in the NSDP and ARCs posted on the SSCU’s website.
  - Disseminating the Data Template on International Reserves and Foreign Currency Liquidity and the IIP.
  - Establishing an interagency committee to determine assignment of roles/responsibilities for quarterly external debt data compilation and dissemination.
  - Disseminating volume measures and index series in relation to a fixed reference base.
- Conduct users’ surveys on a regular basis to review relevance and utility of statistics.
- Provide advance notice of major changes in methodology, source data, and statistical techniques.

### Staff’s recommendations — National Accounts
High priority
- Further revamp data collection and production environment: greater use of sample surveys and improved data flow management and processing.
- Revive the high powered Statistical Council to guide quality and improvement strategies.
- Strengthen SSCU independence to prevent undue official influence on methodological decisions, dissemination practices, and revision policies.
- Compile seasonally adjusted measures of GDP.

Other key recommendations
- Compile fourth quarter national accounts from source data consistent with source data of first three quarters.
- Improve timeliness of fourth quarter national accounts to meet SDDS requirements.

### Staff’s recommendations — CPI and PPI
High priority
- For the CPI: implement procedures in regional offices to impute prices of each missing variety month-to-month until the variety is deemed permanently missing; particularly relevant for seasonally available items (fresh fruits and vegetables).

Other key recommendations
- For the PPI: incorporate more market service-producing activities over time.

### Staff’s recommendations — Government Finance Statistics
High priority
- Re-disseminate annual data for general government for 1999–2001 based on the format for submission to the IMF GFS Yearbook; include nonbudget pension and social security funds.
- Publish more extensive documentation of definitions, sources, and methods to enhance accessibility.
- Incorporate nonbudget social security funds in disseminated central government operations data and improve timeliness.

Other key recommendations
- Set target dates for development plan stages for migration to the 2001 GFS Manual and prioritize implementation including resource allocation.
- Review redeploying resources to implement the development plan and cope with increased workload from bringing local government into the TSA.

### Staff’s recommendations — Monetary Statistics
High priority
- Improve timeliness of analytical accounts of the central bank to meet SDDS using daily accounting records excluding internal organizations.
- Disseminate information on compilation of monetary data, including major aggregates and differences from internationally accepted standards.
- Establish a migration plan to adopt MFSM 2000 recommendations.

Other key recommendations
- Expand coverage to include nonbank financial institutions issuing deposit substitutes and compile/disseminate flow data; increase human and budgetary resources.
- Limit changes in charts of accounts of NBU and commercial banks to twice a year and prepare backward revisions to enhance consistency.
- Separately identify information on financial derivatives.
- Verify consistency of monetary statistics and government finance statistics regularly.

### Staff’s recommendations — Balance of Payments
High priority
- Implement a lower reporting threshold for banks to provide payment order information for high-value transactions.
- Implement a travelers’ survey to improve quality of shuttle trade, compensation of employees, and travel estimates.
- Expand the survey form on inter-enterprise arrears to include financial instruments, currency composition, debt maturity, and write-offs of bad debts; reduce response burden by conducting survey quarterly and adopting sampling techniques.

Other key recommendations
- Increase component details of current transfers.
- Review the FDI survey to record inter-enterprises’ direct investment loans under FDI instead of under other investment.
- Prepare and regularly publish revision studies explaining sources and magnitude of revisions and the impact of new data sources.

*Source: IMF staff report (excerpts), Ukraine data quality assessment.*

### 1. Integrity

### _cr03256 - 1. Integrity

### 1. Integrity — Assessment matrix (selected indicators)
- 1.1 Professionalism: O O O O O O — See paragraph 15.
- 1.2 Transparency: O LO LO LO LO LO — See paragraphs 15, 16, 17.
- 1.3 Ethical standards: O O O O O O.

- 2. Methodological soundness
  - 2.1 Concepts and definitions: O O O O O O.
  - 2.2 Scope: LO LO LO LO LO LO — See paragraphs 18, 19, 20, 21, 22.
  - 2.3 Classification/sectorization: O O O LO O LO — See paragraphs 20, 22.
  - 2.4 Basis for recording: O O O O O O.

- 3. Accuracy and Reliability
  - 3.1 Source data: O O O O O LO — See paragraph 27.
  - 3.2 Statistical techniques: LO O O O O O — See paragraph 23.
  - 3.3 Assessment and validations of source: O LO LO O O O — See paragraph 24.
  - 3.4 Assessment and validation of: O O O O O O.
  - 3.5 Revision studies: LO O NA NA LO — See paragraphs 23, 25, 26, 27.

- 4. Serviceability
  - 4.1 Relevance: LO LO LO LO O LO — See paragraphs 28, 29, 30, 32.
  - 4.2 Timeliness and periodicity: LO O O LO LO O — See paragraphs 28, 30, 31.
  - 4.3 Consistency: LNO LO O LNO LO LO — See paragraphs 28, 29, 30, 31, 32.
  - 4.4 Revision policy and practice: O LO LO O O LO — See paragraphs 29, 32.

- 5. Accessibility
  - 5.1 Data accessibility: LO LO LO LO LO LO — See paragraphs 33, 34, 35, 36.
  - 5.2 Metadata accessibility: O LO LO LO LO O — See paragraphs 33, 34, 35.
  - 5.3 Assistance to users: O O O LO O O — See paragraph 34.

### Overview — context and commitments
- IMF ROSC mission to Ukraine: 3-17 April, 2002.
- Ukraine subscribed to SDDS in January 2003.
- Authorities (SSCU, MoF, NBU) agree with ROSC conclusions and view IMF analysis as a basis for further development.
- Emphasis on: improving compilation and dissemination, strengthening coordination among SSCU, MoF, NBU, and enhancing public provision of statistics and metadata via publications, press releases, and the Internet.

### II. Response to Staff’s Recommendations — General
- NSDP and ARCs: NSDP posted and regularly updated on the SSCU web-site; ARC information updated regularly.
- Data Template on International Reserves and Foreign Currency Liquidity: compiled in line with SDDS and disseminated on the NBU web-site.
- Interagency committee for quarterly external debt data: NBU and MoF responsibility to be fixed after government approves relevant legislation acts.
- Volume measures and index series: National Accounts volume measures, Industrial Production Іndex, CPI and PPI Іndex series recalculated to a fixed reference period (2001 =100).
- Users’ surveys: Plan to be part of project “Ukraine Statistics. Strategy and Implementation Plan. 2004-2008” (being developed by SSCU).
- Advance notice of major changes: Information on methodology/methods to be presented in statistical publications, posted on SSCU web-site, and provided to the IMF.

### National Accounts — priorities and actions
- High priority actions
  - Greater use of sample surveys and improved data flow management: first stage of statistical register introduced; work on statistical indicators’ database started.
  - Revive the Statistical Council: included in the “Ukraine Statistics. Strategy and Implementation Plan. 2004-2008”.
  - Strengthen SSCU independence: Statistical Council to advise on methodological approaches and ensure adoption of internationally recognized scientific principles.
  - Compile seasonally adjusted measures of GDP: introduction of quarterly indicators at base-period prices and accumulation of sufficient index numbers (national classifications) to allow seasonal adjustment.
- Other key recommendations
  - Compile fourth quarter national accounts from source data consistent with first three quarters: implemented for fourth quarter of 2002 per State Plan of Statistical Works for 2003.
  - Improve timeliness of fourth quarter national accounts to meet SDDS: fourth quarter national accounts to be compiled on the 90th day after the reference quarter.

### Consumer Price Index (CPI) and Producer Price Index (PPI)
- CPI — High priority
  - Implement regional procedures to impute prices for each missing variety month-to-month (especially seasonal items like fresh fruits and vegetables): experimental calculations of imputed prices for missing seasonal items are being made at regional level; decision on implementation pending based on results.
- PPI — Other key recommendations
  - Incorporate more market service-producing activities over time: issue under consideration.

### Government Finance Statistics (GFS)
- High priority
  - Re-disseminate annual general government data for 1999–2001 in IMF GFS Yearbook format, including nonbudget pension and social security funds: MoF sent letter No025-06/52 dated September 17, 2002 to IMF Statistics Department with material for IMF Yearbook; IMF comments incorporated for 2001 and previous year data provided.
  - Publish more extensive documentation of definitions, sources, methods: data placed in Budget of Ukraine – 2001 national statistics yearbook; list of definitions posted on MoF web-site.
  - Incorporate nonbudget social security funds in central government operations and improve timeliness: MoF revising accounting forms of social insurance funds and Pension Fund in 2003 for consolidation for statistical purposes.
- Other key recommendations
  - Set target dates and allocate resources for migration to the 2001 GFS Manual: MoF developed strategy for reforming government finance statistics for 2003-2006 and passed action list to SSCU (part of World Bank Project on Reform the State Statistics System).
  - Treasury servicing of local budgets: Treasury local budgets servicing started on January 1, 2002; stage-by-stage introduction of Treasury servicing of local budget disbursement planned up to 2005.

### Monetary Statistics
- High priority
  - Improve timeliness of analytical accounts of the central bank to meet SDDS using daily accounting records excluding internal organizations: analytical accounts to be reported using a flexibility option no later than the 30th day after the end of the reference month within January–November and within 6 weeks after the end of the reference year because of end-of-year delays.
  - Disseminate information on compilation of monetary data and differences from international standards: monetary data compiled in accordance with IMF SDDS requirements; posted on NBU web-site and published in monthly “Bulletin” of the NBU.
  - Establish migration plan to adopt MFSM 2000 recommendations: preparation work continued; intend to expand coverage to include non-bank financial institutions and derivative financial instruments.
- Other key recommendations
  - Expand coverage to include nonbank financial institutions issuing deposit substitutes and compile/disseminate flow data; increase resources: practicable after establishing supervisory body on non-bank financial institutions.
  - Limit changes in charts of accounts and prepare backward revisions to enhance consistency: Resolution of the National bank of Ukraine No124 of 19.03.2003 determined that changes in charts of accounts shall be made once a quarter.
  - Separately identify information on financial derivatives: Full information on financial derivatives is provided in accordance with their usage.
  - Verify consistency of monetary statistics and government finance statistics on a regular basis: consistency verified regularly.

### Balance of Payments
- High priority
  - Implement lower reporting threshold for banks on payment orders for high-value transactions: NBU intends in 2003 to reduce reporting threshold in ITRS from 250 to 50 thousands U.S. dollars; changes approved by Resolution of the Board of the National Bank of Ukraine No 124 of 19.03.2003.
  - Implement travelers’ survey to improve shuttle trade, compensation of employees, and travel estimates: Cabinet and NBU resolution provides for using joint travellers' survey data (State Customs Service, SSCU, NBU) when compiling BOP; extra financing for survey not yet solved.
  - Expand inter-enterprise arrears survey form to include financial instruments, currency composition, debt maturity, write-offs; consider quarterly frequency and sampling to reduce response burden: NBU recognizes need to modify enterprise survey to capture these items.
- Other key recommendations
  - Increase component details of current transfers: ITRS improvements to allow more detailed comments on transactions, strengthening control on respondent information quality.
  - Review FDI survey to record inter-enterprise direct investment loans as FDI rather than other investment: new Form 10 “Report on Foreign Investments in Ukraine” developed in 2002; implementation in 2003 expected to allow identification of credits of direct foreign investors and record them under “Direct investment”.
  - Prepare and regularly publish revision studies explaining sources and magnitude of revisions: information on revisions to be published in NBU Balance of Payments official publication together with 2002 BOP data.

### DQAF and Documentation
- Document purpose: detailed assessment by dataset of elements and indicators underlying the DQAF used in the ROSC—Data Module; includes generic framework and users’ survey results as appendices.
- Contents listing (sections): I. National Accounts Statistics; II. Price Statistics (CPI); III. Price Statistics (PPI); IV. Government Finance Statistics; V. Monetary Statistics; VI. Balance of Payments.

*Prepared by the Statistics Department; Approved by Carol S. Carson and John Odling-Smee; August 5, 2003.*

### 1. DQAF: Summary of Results for National Account Statistics..............................................24

### 1. DQAF: Summary of Results for National Account Statistics

### I. Prerequisites of quality — Legal and institutional environment
- The Statistics Law of Ukraine adopted July 13, 2000 (No. 1922–III) and effective January 1, 2001 defines roles and responsibilities of state statistical agencies and is supplemented by Presidential Decree April 14, 1995 (No. 312/95) establishing the SSCU as the central executive agency for official statistics.
- Key regulatory acts and dates establishing national accounts responsibilities:
  - Resolution of the Cabinet of Ministers December 28, 1992 (No. 727).
  - Resolution of May 4, 1993 (No. 326) specifying preparation of annual and quarterly GDP estimates.
  - Presidential Decree November 6, 1997 (No. 1249/97) making national accounts a routine SSCU task and delegating internal organization powers to the SSCU chairperson.
  - Cabinet of Ministers order August 9, 2001 (No. 341–r) establishing periodicity and timeliness of GDP estimates and strengthening inter-agency data provision schedules.
- Medium-term reform program: Cabinet of Ministers resolution June 27, 1998 (No. 971) (program for 1998–2002); a longer-term 10-year program was under preparation.

### I. Prerequisites of quality — Data sharing and coordination
- Legislative basis: Article 7 of the Statistics Law entitles SSCU to use administrative, banking, financial, and balance of payments data from other agencies.
- Specified data provision deadlines in Cabinet order (No. 341–r) include:
  - NBU annual commercial bank interest income/interest payments: within 30 days after end of accounting year.
  - NBU monthly data: within 20 days after the end of the reporting month.
  - Annual profit and loss statements of commercial banks and NBU: by end of May following the reporting year.
  - Quarterly profit and loss statements: by the twenty-fifth day following the end of the reporting period.
  - Annual balance sheets of NBU and commercial banks: by end of September following the reporting period.
  - Quarterly balance sheets: within 75 days after the end of the reporting quarter.
  - Preliminary annual balance of payments data: within 40 days after the end of the reporting year.
  - Quarterly balance of payments data: within 75 days after the end of the reporting quarter.
- Protocols and inter-agency working groups formalize coordination (examples: Tax Inspection protocol February 18, 1997; Ministry of Economy protocol February/March 2000; Ministry of Transport protocol June 2001).
- Inter-agency working group coordinated by SSCU (including MoF, STU, NBU, research agencies, Cabinet of Ministers) established since September 25, 2001.

### I. Prerequisites of quality — Confidentiality and respondent obligations
- Articles 21 and 22 of the Statistics Law prohibit dissemination of individual/confidential statistical information without respondent permission and forbid state/local authorities or others from obtaining such information.
- Article 17 requires staff to protect confidential information; access limited to official purposes.
- Sanctions for disclosure under law on Administrative Violations (as amended July 13, 2000 No. 1929–III):
  - Fines for individuals: 3–5 times the pre-tax minimum personal income.
  - Fines for registered economic units and entrepreneurs: 10 to 15 times the pre-tax minimum personal income.
  - Higher fines for repeated violations.
- Article 18 entitles respondents to know nature, purpose, and ultimate use of collected data.
- Article 13 authorizes SSCU to obtain required data (including accounting reports) and to check status and reliability; Article 18 obliges respondents to provide required data free to SSCU.
- Penalties for nonresponse, false, inaccurate, or untimely information mirror disclosure fines: 3–5 times pre-tax minimum personal income for individuals; 10–15 times pre-tax minimum personal income for legal units and entrepreneurs.

### I. Resources — Staffing, financial, and computing resources
- Department of National Accounts organizational structure: divisions for national accounts, sub-annual national accounts, consumption and capital formation accounts, production and income accounts, and supply-use tables.
- Staffing:
  - Total staff in these divisions: 35, all university educated in statistics, economics, information technology, and accounting.
  - Senior staff experienced in 1993 SNA and international IMF courses; turnover to NBU and other agencies noted.
- Assessments:
  - Current staff, financial, and computing resources are adequate for compiling the current national accounts series.
  - Additional staff required for timely development of full institutional sector accounts, including capital and financial accounts.
  - Current staffing does not allow annual compilation of comprehensive supply-use tables in current and constant prices.
- Computing:
  - Staff have computers connected to SSCU’s local area network, but several machines are of an old vintage, limiting adoption of state-of-the-art processing techniques.
  - Specialized computing training would yield efficiency gains.

### I. Resources — Measures to ensure efficient use of resources and organizational arrangements
- Recent SSCU organizational adjustments:
  - Staff reduced from more than 21,000 to approximately 12,800; headquarters staff approximately 440 in Kyiv.
  - Central Inter-Regional Department of Statistics (main Computer Center) employs an additional 300 staff.
  - Regional network: 25 oblast offices, 2 city offices (Kyiv and Sevastopol), and 696 sub-offices in rayons.
- Data collection and processing practices:
  - Continued reliance on full or very large counts in many sectors; sample surveys introduced in some sectors but not universally.
  - Data processing and editing occur at multiple levels (rayon → oblast/city → main Computer Center), resulting in duplication: micro-level and aggregated data submitted to main Computer Center and rechecked at multiple stages.
  - Recommendation implicit in assessment: accelerate implementation of sample surveys, improve data flow management, and centralize processing of primary data in the main Computer Center in Kyiv to streamline procedures.
- Human resources management:
  - Staff performance appraisal follows civil service law; appraisals every three years with departmental managers adjusting bonus portions typically once a year.
  - Department of Finance and Computation monitors staff utilization; regional workload monitored informally via communication with regional managers.
  - Compensation comprises base salary, performance-based bonus, and cash awards; appointment of new staff is competitive and by order of the SSCU chairperson.

*Summary extracted from "Detailed Assessment Using the Data Quality Assessment Framework (DQAF)"—National Accounts Statistics section.*

### 0.3 Quality awareness

### 0.3 Quality awareness

### Processes to focus on quality
- Management encourages adoption of procedures for quality checks during data collection, processing, and compilation (see 3.3 and 3.4, and footnote 5).
- Management is aware of efficiencies from centralizing data processing in the main Computer Center in Kyiv; two new report forms for nonfinancial enterprises (1-predprinimatelstvo and balance sheet information of enterprises) are processed in the main Computer Center. These forms were recently “unified” by combining several old report forms.
- The annual turnover of staff is about 10 percent.

### Monitoring the quality of collection, processing, and dissemination
- The system of national accounts is viewed as the conceptual framework for economic statistics (see 3.3.1).
- Reduced format supply-use tables are compiled annually since 1999, mainly for analytical purposes to identify major discrepancies (see 4.3.2).
- Regional publications are reviewed quarterly by headquarters staff with recommendations on accuracy, format, and presentation (including use of graphics). Annual reviews make recommendations on quantitative and descriptive content.
- There is currently no body distinct from the SSCU that provides guidance on the quality of statistical series and strategies for improving data production.
- A high-powered Statistical Council attached to the office of the President, that existed some time ago, could be revived for this purpose. The Council could play an active advisory role and, when required, may examine issues in depth and make specific and detailed recommendations to the chairperson of the SSCU, in particular, regarding methodological issues, accuracy, and exhaustiveness of official statistics. It could be chaired by the chairperson of the SSCU and comprise representatives of enterprises, trade unions, ministries, regional statistical offices, academic/research institutions, etc., and meet as required.
- The Presidential decree of November 6, 1997 (No. 1249/97) allowed the establishment of an Advisory Council comprising the chairperson, deputy chairpersons, other senior officials of the SSCU, as well other persons approved by the Cabinet of Ministers.
- Annual and sub-annual GDP estimates have an active user base inside and outside government; National Accounts Department staff receive several queries every month related to GDP issues and undertake periodic consultations with users supplemented by informal contact (see 4.1.1).

### Dealing with quality trade-offs and planning for needs
- Management recognizes trade-offs among dimensions of quality, especially when meeting users’ demand for high-frequency national accounts data.
- To economize resources, management recognizes the need to:
  - reduce demand for detailed administrative data of local rayon, oblast, and city administrations;
  - reduce the requirement to produce timely estimates of monthly GDP.
- Management is aware that quality could be improved by reorienting the data collection effort within sampling frameworks (see also 0.2.2 footnote 5).

### Key source-data and coverage facts relevant to quality
- A comprehensive statistical register derived from the Unified State Register of Enterprises and Organizations (USREO) has recently become operational with assistance from Eurostat.
- The statistical register includes around 450,000 registered units classified using the new five-digit national industrial classification (see 2.3.1). Summary information (in Ukrainian and English) from the register is published regularly. A special survey of 10,000 statistical units was conducted in 2000 to update the statistical register. The register is also routinely updated using information provided by enumerators, and by the tax register.
- In 1999 the surveys supporting the national accounts program covered 88.9 percent of total output.
  - Data for compiling GDP by economic activities came from almost 37,000 medium and large enterprises and 197,000 small enterprises.
  - Sectoral coverage in 1999: mining 99.5 percent, electricity, gas, and water production 92.8 percent, manufacturing 92.8 percent, agriculture and forestry 96.2 percent, wholesale and retail trade 51.9 percent, services 67.9 percent.
- In the latest annual data collection cycle of 2001:
  - number of medium and large enterprises covered increased to 47,489;
  - number of enterprises covered in the quarterly surveys ranged from 46,248 to 47,489.
- National accounts data for enterprises are collected mainly using two report forms: one for production and production-related expenses (including changes in inventories) (1-predprinimatelstvo) and another for financial and balance sheet data.
- Government finance statistics are obtained on a monthly and annual basis from the STU and allow measurement of output, intermediate consumption, fixed capital formation, and final consumption expenditure of government.
- The first all-Ukrainian population census was conducted in 2001; it will be conducted once in 10 years.
- The new quarterly household survey—conducted since 1999—has comprehensive coverage, sound sample design, and editing and imputation procedures; raising factors are derived from the sample design. The surveys provide detailed information on expenditure and sources of income. The data are compiled within four to four and a half months after the end of the reference period.
- The labor force survey is conducted on a quarterly basis since 1999; prior to that it was annual.

*Source: _cr03256 - 0.3 Quality awareness*

### 3.2 Statistical techniques

### 3.2 Statistical techniques

### Production approach — compilation and volume measures
- Annual and quarterly estimates of gross output are compiled for most sectors at the two-digit level of the national classification from data reported by enterprises; intermediate consumption estimates are compiled at the same level of detail. Data dissemination: 38 economic activities for annual estimates, and 10 economic activities for quarterly estimates.
- Gross output of owner-occupied dwellings is valued as the estimated rentals that tenants would pay for similar accommodation.
- The concept of work-in-progress is applied to growing crops, standing timber, stocks of fish, livestock reared for purposes of food, large construction projects, and output of large equipment.
- Consumption of fixed capital estimates are derived from data reported by enterprises on capital amortization.
- Government subsidies, revenue (excluding taxes), and expenditure are on a cash basis and are not converted to an accruals basis.
- Changes in inventories are estimated accounting for holding gains/losses. The GDP estimate for 2001 includes an adjustment for holding gains/losses of around 2 percent of GDP.
- Volume measures are compiled at a disaggregated level and gross output is extrapolated using production indices/volume data for agriculture, mining, manufacturing, trade, construction, market and nonmarket services, etc.
- Agricultural volume index: monthly Laspeyres-type index with 2000 as the base year; data for compiling the index come from area surveys, census of livestock, and productivity data derived from spot surveys of agricultural production in rural areas (covering 0.44 percent of total cropped area).
- Manufacturing volume index: compiled from data reported directly by enterprises in current prices and in constant prices of December 2000 (facilitates direct calculation of volume growth rates); uncertainty remains about how enterprises arrive at constant price data and how they deal with new products or varieties.
- Construction current price output is deflated using a construction price index with base 1991; the index is compiled from information on 92 cost item groups as reported by selected enterprises (contractors).
- Transportation gross output is extrapolated using volume indices compiled from freight volume and number of passengers.
- Trade: previous period current price gross output of trade and restaurants extrapolated using index of aggregate sales of consumer goods; index compiled by deflating actual sales with corresponding sub-components of the retail price index (retail trade) and producer price index (wholesale trade). Trade margins output volume estimated by extrapolating base-year trade margins using volume extrapolators of sales (retail and wholesale).
- Market services (including social and cultural services) are deflated using corresponding CPI sub-components. Double deflation is applied to housing and communal services using aggregate supply-use tables.
- Science-related services, finance services, and nonmarket services are extrapolated using the employment index of the corresponding sectors.
- Net taxes and subsidies are extrapolated using volume indices of the respective sectors.
- SSCU plans to improve methodology: compile better volume indices (e.g., index of manufacturing production by 410 manufacturing groups and 13 main sectors) and use alternative methods such as deflation of values using producer price indices at the detailed level.10

### Expenditure approach — compilation and limitations
- GDP by expenditure components are derived independently.
- Household final consumption expenditure compiled by 12 main groups corresponding to COICOP.
- Government consumption expenditure and gross fixed capital formation compiled following COFOG.
- Expenses of residents abroad are not fully covered in household final consumption expenditure due to insufficient data.
- Gross fixed capital formation broken down by machinery and equipment, construction and assembly, capital repairs, etc.
- Changes in inventories distinguished by three main types: raw material, work-in-progress, and finished goods; compiled from data reported directly by enterprises.
- Price/volume indices used at a fairly disaggregated level to derive volume measures:
  - Household final consumption deflated using CPI of broad groups.
  - Government final consumption extrapolated using employment data.
  - Gross fixed capital formation deflated using price index of construction machinery.
  - Equipment deflated using PPI for the corresponding group.
- Lack of proper quarterly export and import price indices reduces the quality of the volume measures of GDP by the expenditure approach.

### Specific quarterly compilation techniques and issues
- No discrepancy between the preliminary annual estimate and the sum of quarterly estimates because the fourth quarter estimate is derived as a residual, after revising the first three quarters; this makes the fourth quarter estimate inconsistent with estimates of the first three quarters.11
- Annual compilation covers a much larger scope than quarterly compilations (annual includes small enterprises that quarterly does not).
- Quarterly data are not seasonally adjusted while the data indicate strong seasonality.
- Improvements in methodology now allow compilation of fourth quarter estimates that are consistent with source data of the three previous quarters.11

### Other statistical procedures — addressing under-recording and informal activities
- Prevailing view that private activities are under-recorded; an estimate of unrecorded activities is included in GDP derived mainly from reported trade turnover.
- Official estimate of unrecorded activities in 2000 is 20 percent of GDP.
- An estimate of the size of the informal sector is also calculated.
- Most unrecorded activities are in construction, trade and transportation, food and light industry, medical, and education services.
- Construction unrecorded activities estimate compiled from an experimental survey of garages, dachas, and private houses in villages.
- Retail trade data source: special survey of specific markets (since 1996) to collect information on sales of specific agricultural products.
- Transportation survey conducted since 2001 mainly to collect data on intra-city and inter-city transportation.
- Commodity balances used to develop aggregate production estimates for specific products (e.g., grain, meat, milk) by comparing production surveys with trade turnover and/or household consumption data and census of livestock.

### Assessment and validation of source data and intermediate data
- Information not available about sampling/nonsampling errors of surveys; nonresponse is followed up and appropriate imputations are made.
- Report forms are reviewed by management and Methodology Division staff, who cooperate with National Accounts Department to ensure consistency of concepts and definitions across survey areas.
- Control over source data is exercised through data checks (example: seven controls for manufacturing production, sales, employment, and expenditure on production). Information collected in report forms 1-Predprinimatelstvo and financial report form for enterprises; appropriate data aggregated and compared.
- High value transactions are given attention.
- Administrative data, budgetary data, international trade data, and other data used in calculations are routinely assessed for deviations from trend and for errors.
- Main intermediate data are validated against other primary/secondary sources where available; deviations in data sources are investigated and adjusted accordingly.
- Statistical discrepancy between GDP by production activities and GDP by expenditure components is investigated (also see 4.3.1).

### Revision studies and practices
- Revision to the final GDP estimate is generally small; procedures for revising the data are not explained.
- Quarterly and annual data are revised once according to a predetermined schedule.
- The revision period of one year for national accounts does not permit incorporation of the latest balance of payments statistics finalized after 15 months.
- Quarterly national accounts are unable to use the latest household survey results prepared four months after the end of the reference quarter (see 3.1.1).
- Preliminary data are clearly identified; revised data are disseminated with the same detail as preliminary data.
- In cases of major revisions, the sources of data revision are indicated to show differences from preliminary figures.

### Serviceability — relevance, timeliness, periodicity, and consistency
- Users’ surveys are not conducted regularly; feedback obtained via working groups and correspondence from official users (MoF, Ministry of Economy, other ministries, research organizations).
- SSCU provides detailed administrative data to oblast, rayon, and city administrations upon request.
- Management intends to refine measurement of unrecorded activities and to compile and disseminate comprehensive supply-use tables; management is aware of need to compile and disseminate consistent time-series data and volume measures in relation to a fixed reference base.
- Timeliness:
  - Timeliness of quarterly data for the first three quarters: 95 days.
  - Fourth quarter data disseminated around four months after the end of the reference period (need improvement to meet SDDS requirements).
  - Preliminary annual data released within two months of the end of the reference period.
  - Final annual data released within one year.
- Periodicity: quarterly and annual data are compiled; rough monthly estimates of GDP are also compiled.12
- Consistency:
  - Statistical discrepancy between GDP by production and by expenditure is usually less than 2 percent of GDP (before adjustment); discrepancy not shown explicitly and removed through adjustments in weaker-source estimates.
  - Concepts and definitions for quarterly and annual GDP estimates are identical (see 3.2.1 on quarterly compilation techniques).
  - Supply-use tables serve as an analytical check on consistency.
  - Consistent annual time series data for GDP are only available for the period 2000–01.
  - Consistent current price quarterly data are only available from 2001.
  - Time-series data for 1990–2000 are according to the old classification of economic activities.
  - Quarterly deflation methodology leads to quarter-to-quarter volume measures that are not consistent; volume measures expressed in prices of the same quarter of the previous year are inadequate for producing consistent quarterly time series and produce volume measures not consistent with annual volume measures of GDP.
  - Total net lending/net borrowing from national accounts is consistent with balance of payments data; rest of the world accounts compiled from balance of payments. Timing differences in release of revised balance of payments and national accounts can produce differences.
  - Government sector accounts and government finance statistics are compiled from the same data sources; estimates are made to convert taxes and wages from cash to accrual basis.

### Accessibility, metadata, and assistance
- National accounts data published with various levels of disaggregation and corresponding metadata are adequate, but data are not disseminated in time-series format (volume measures not disseminated relative to a fixed reference base).
- Data are catalogued; brief descriptions and notes on changes provided. Annual growth rates and percentage shares are shown.
- Quarterly estimates disseminated in current prices and in prices of the same period of the previous year; data provided by 10 main types of economic activity.
- Preliminary annual data released via press release; final annual data released with maximum detail in a dedicated national accounts publication.
- Data released on a pre-announced schedule and made available to all users at the same time in accordance with the Statistics Law.
- Nonpublished (but nonconfidential) sub-aggregates are available upon request from the national accounts contact person.
- Metadata:
  - Brief documentation provided with quarterly data releases; more details provided in annual publication.
  - Main provisions of statistical methodology are published.
  - Metadata available at different levels: brief notes, summary of methodology, detailed explanation.
- Assistance:
  - A contact person for national accounts is available to provide assistance; information about the contact person is not publicized.
  - Catalogue of SSCU publications is available; prices shown and a contact telephone provided for further information.

*Source: _cr03256 - 3.2 Statistical techniques.*

### 1. Integrity

### 1. Integrity

### Professionalism
- Official users exercise undue influence in determining the dissemination practices and revision policy of national accounts.
- Implementation of the national accounts program relies excessively on government regulatory support.
- Assessment marker: X (Professionalism not fully observed).

### Transparency
- Assessment marker: X (Transparency concerns noted).

### Ethical standards
- Assessment marker: X (Ethical standards not fully observed).

### 2. Methodological soundness

### Concepts and definitions
- Assessment marker: X

### Scope
- Assessment marker: X
- Comment: The scope is observed in principle, but cannot be fully implemented because of source data constraints.

### Classification/sectorization
- Assessment marker: X

### Basis for recording
- Assessment marker: X

### 3. Accuracy and Reliability

### Source data
- Assessment marker: X

### Statistical techniques
- Assessment marker: X
- Comments:
  - Fourth quarter estimates are compiled residually from data for the whole year and the sum of the first three quarters.
  - There is no proper quarterly price indices for exports and imports.
  - Volume index for manufacturing is compiled from data reported directly by enterprises in constant prices.
  - Quarterly GDP is not seasonally adjusted.

### Assessment and validations of source data
- Assessment marker: X

### Assessment and validation of intermediate data and statistical outputs
- Assessment marker: X

### Revision studies
- Assessment marker: X
- Comment: The procedures for revising the data are not explained.

### 4. Serviceability

### Relevance
- Assessment marker: X
- Comment: Formal users’ surveys are not conducted.

### Timeliness and periodicity
- Assessment marker: X
- Comment: Fourth quarter data are disseminated more than three months after the end of the reference quarter.

### Consistency
- Assessment marker: X
- Comment: Consistent current price time-series data are not available for an adequate period of time. Consistent quarter-to-quarter volume measures of GDP are not available.

### Revision policy and practice
- Assessment marker: X

### 5. Accessibility

### Data accessibility
- Assessment marker: X
- Comment: Data are not disseminated in time-series format.

### Metadata accessibility
- Assessment marker: X

### Assistance to users
- Assessment marker: X

*National Accounts Statistics — Data Quality Assessment Framework: Summary of Results for National Account Statistics (Compiling agency: State Statistics Committee).*

### Price Statistics (Consumer Price Index) — Prerequisites of quality

### 0.1 Legal and institutional environment
- The current Statistics Law of Ukraine, adopted by the Parliament on July 13, 2000 (No. 1922–III) and effective January 1, 2001, defines the roles and responsibilities of state statistical agencies.
- Presidential Decree of April 14, 1995 (No. 312/95) establishes the State Statistics Committee of Ukraine (SSCU) as the central executive agency for official statistics.
- Presidential Decree of November 22, 1997 (No. 1299/97) mandated development of an integrated system of statistics according to international standards and directed the Cabinet of Ministers to support the new statistical system.
- Article 12 of the Statistics Law assigns SSCU responsibility to collect, process, aggregate, analyze, disseminate, store, and protect statistical information.
- Price statistics are produced within the SSCU’s Price Statistics Department and do not directly require inter-agency cooperation.
- Legislative basis for protecting respondent confidentiality is strong (Articles 21 and 22). Conflicts with other legislation have arisen but courts have supported the SSCU; amendments have been introduced to remove contradictions.
- Under Article 17 staff must protect confidential information; Article 20 imposes liability for disclosure. Fines for breach:
  - Individuals: 3–5 times the pre-tax minimum personal income.
  - Registered economic units and entrepreneurs: 10–15 times the pre-tax minimum personal income.
  - Fines are higher for repeated violations.
- CPI data comprise publicly observable retail prices; weights derive from Household Living Conditions Survey (HLCS) data.
- PPI microdata and calculation systems are protected with passwords and additional physical safeguards.
- Article 18 entitles respondents to know what primary data are collected, for what purpose, and who will use them.
- Article 13 authorizes SSCU to obtain required data, including accounting reports and explanations; Article 18 obliges respondents to provide required data free to the SSCU.
- Penalties for nonresponse, false, inaccurate, or untimely information:
  - Individuals: fines amounting to 3–5 times the pre-tax minimum personal income.
  - Legal units and entrepreneurs: fines of 10–15 times the pre-tax minimum personal income.
  - In practice, punishments are rarely applied.

### 0.2 Resources
- Overall numbers of staff appear adequate.
- Budgeted staff positions for combined CPI, PPI, and construction price index programs (SSCU):
  - All price program positions: 570
  - Headquarters (Price Statistics Department): 21 (CPI 6; PPI 5; Construction Price Index 2; Retail Price Index & Communication & Transport Price Indices 6; Administration 2)
  - Computer Center: 9 (CPI ---; PPI 6; Construction Price Index 1; Retail Price Index & Communication & Transport Price Indices 2; Administration ---)
  - Regions: 540 (Centers 308; Districts 232)
- Current budgeting tracks staff allocation by program at headquarters and Computer Center, but not in regions where staff may work on multiple programs.
- Financial resources have improved recently; SSCU also earns income from subordinated bodies through special tabulations, publications sales, computer and research services, and specialized training.
- Computing resources in headquarters: local area network of Pentium personal computers with standard word processing and spreadsheet software; used to compile CPI from 270 item indices transmitted from each of the 27 regions each month.
- Budget limited for hardware and software upgrades; maintenance handled by SSCU’s Department of Computation.
- Receipt of CPI and PPI data from regions via electronic mail to the Computer Center in Kyiv; PPI microdata and calculation maintained centrally.
- Regional computing resources are barely adequate; PPI data entry uses custom Clipper application on older systems; PPI calculation at Computer Center uses another custom Clipper program.

### 0.2.2 Measures to ensure efficient use of resources
- Department of Finance and Department of Computation maintain staff utilization data for each SSCU department at headquarters, Computer Center, and regions.
- Managers of Price Statistics Department do not routinely use these data due to lack of regional program detail; regional workload monitored informally via communication.
- Staff compensation comprises base salary, performance-based bonus, and cash awards.
- Appraisals occur at three-year intervals; bonus adjustments normally once a year but can be made at any time.

### 0.3 Quality awareness
- No formal quality program exists currently; management encourages procedures for quality checks during data collection, processing, and compilation.
- Ongoing quality improvement projects as part of Statistical Reform Program (1998–2002) and Statistics Development Program (2003–2007).
- CPI improvement project (2000–01):
  - Reduced sample from 475,000 price records per month to 275,000 without harming national sample coverage.
  - Implemented COICOP/HBS classification.
- PPI implementing Eurostat’s NACE industry classification and CPA product classification in the first half of 2002; compiled with improved central database application.
- CPI and PPI data models derive weights from same household expenditure survey and industrial survey sources used in national accounts; reconciliations between national accounts and price indices are not viewed as worth undertaking.
- Time-series consistency is examined monthly between comparable components of CPI and PPI.
- Processes to monitor quality:
  - Report forms reviewed by management and Methodology Division; Methodology Division tests reports with small group of enterprises.
  - Input-output table compilation serves as a consistency check for national accounts.
  - Unusual movements in price data are reviewed at regional and central levels and traced to reporters if necessary.
  - For PPI, personal visits are made twice a year to verify responses received principally by mail.
  - Regional office error performance is routinely tracked; problems addressed with regional managers.
  - SSCU’s General Information Department checks all publications for accuracy; quarterly and annual reviews of regional publications undertaken.
- No external body distinct from SSCU provides guidance on series quality; a high-powered Statistical Council attached to the office of the President could be revived.
- Government policy users have been surveyed under Presidential Regulation on the Observation of Prices and Tariffs (2001–03); no formal surveys of nongovernment users.
- SSCU’s Price Department reports routinely receiving 40-50 letters a month on price programs and received 543 letters on the CPI during 2001.
- SSCU’s website generates large volumes of electronic mail and frequent telephone inquiries.
- Staff report that timeliness requirements (release by the tenth of the month following the reference month) do not conflict with quality; checks are completed within available processing and review window.

*Source: _cr03256 - 1. Integrity*

### 1. Integrity

### 1. Integrity

### 1.1 Professionalism
- Statistics are compiled on an impartial basis:
  - Article 5 of the Statistics Law ensures professional independence of the SSCU as the central executive body for statistics and forbids interference by State and local authorities, officials, public associations, and other persons.
  - SSCU is an independent organization with its own set of financial accounts and identifiable seal.
  - Presidential Decree of November 6, 1997 (No. 1249/97): chairperson of the SSCU is appointed by the President on recommendation of the Prime Minister, reports to the President and the Cabinet of Ministers, and is responsible for ensuring professionalism in statistical activities.
  - Civil service law determines staff recruitment qualifications; staff compensation is determined by individual accomplishment in meeting objectives of the statistical work program; professional skills in statistics and economics are considered in compensation and promotion decisions.

- Choices of sources and statistical techniques are informed solely by statistical considerations:
  - Articles 5, 7, 8, and 9 of the Statistics Law guarantee independence in selection of methodology, data sources, content, and format of data disseminated by SSCU.
  - SSCU’s Price Department reports no external interference in its work.

- Right to comment on erroneous interpretation and misuse:
  - Article 13 of the Statistics Law empowers SSCU to comment on misuse or misinterpretation of statistical information.
  - SSCU may communicate disagreements in print, other public modalities, or direct private communication with issuers/authors.

### 1.2 Transparency
- Terms and conditions of data collection, processing, and dissemination:
  - The Statistics Law is public and available on the SSCU official website.
  - Statistical publications provide contact information for further inquiries.

- Internal governmental access prior to release:
  - Article 14 guarantees equal access to all data users; data are released simultaneously to all interested parties.
  - SSCU and its Price Statistics Department provide no access to price statistics before official release.

- Identification of products:
  - All government publications of SSCU data must properly attribute them to SSCU.

- Advance notice of methodological changes:
  - Advance notice provided in statistical development programs and the annual plan of the SSCU (public and provided to major users).
  - Users are informed via notes in the CPI methodological manual and annual bulletin after implementation.
  - No additional advance notice is given in the monthly Express Report; recommendation: include an additional alert in the Express Report just prior to implementation to advise users in advance.

### 1.3 Ethical standards
- Staff behavior and confidentiality:
  - Article 17 of the Statistics Law requires all employees of state statistical bodies to protect confidential information.
  - Civil Service Code provides penalties for misuse of public property and confidential information.
  - Staff are informed by management and must commit to these regulations as a condition of employment.

### 2. Methodological soundness

### 2.1 Concepts and definitions
- Overall structure:
  - CPI concepts and definitions fit within the requirements of the 1993 SNA and the guidance in Consumer Price Indices, An ILO Manual (1989).

### 2.2 Scope
- Scope consistency:
  - Ukrainian CPI broadly follows international practice, including almost all monetary consumption expenditure but excluding transactions in used goods.
  - No nonmonetary expenditure is covered.

### 2.3 Classification/sectorization
- Classification systems:
  - Ukrainian CPI employs Eurostat COICOP/HBS expenditure classification derived from COICOP.
  - Institutional sector coverage follows the 1993 SNA household sector with conventional exclusions: households residing in rural areas, persons living on military bases, residents of hospitals and prisons.
  - CPI covers the urban population, which comprises about 68 percent of the total as of 2001.
  - Population size measure for actual CPI sample of areas is based on the 1989 census updated to 2000 with vital statistics; urban population estimate for 2000 also was 68 percent of the total.

### 2.4 Basis for recording
- Market prices:
  - CPI weights and prices use market (actual transaction) prices.
  - Detailed product specifications attempt to include most important price-determining characteristics; tracking in computerized databases varies by regional office computer resources.

- Accrual recording:
  - Weights and base prices cover the full reference year; monthly price observations distributed over the month from the first through the twenty-fifth.
  - Weights are based on obligations households accrue (accrual basis), incorporate adjustments for payment arrears (e.g., communal utility services), and include full purchase value of items acquired on credit.

- Grossing/netting:
  - Not applicable because used durable goods and owner-occupied housing are excluded.

### 3. Accuracy and Reliability

### 3.1 Source data
- CPI weights source:
  - Source is the HLCS: comprehensive, quarterly, timely (eight-month processing lag), employs international best practice.
  - Sampling frame based on the most recent available population census (1989), updated by vital statistics on births and deaths.
  - HLCS includes information for area sampling (addresses, income, household size) and uses a modern, multistage randomized area sample design.

- CPI prices source and sampling frames:
  - First-stage CPI price sample: districts (rayons) selected within each of the 26 regions (oblasts) other than Kyiv (Kyiv comprises a single district); area frame stratified by population size before district selection.
  - Total consumption expenditure would be preferable for stratification, but district-level expenditure data are not available; population is used as a proxy.
  - No sampling frames available for second-stage (retail outlets for 270 item groups) and third-stage (elementary items) selections.
  - Business register for retail enterprises could be a medium-term option for outlet sampling if kept current; would require supplementary canvasses and post-stratification by product line.
  - Recent streamlining reduced CPI sample by almost 60 percent to about 23,000 outlets and about 275,000 price records per month, potentially enabling probabilistic disaggregation for selecting elementary items as a medium-term option.
  - HLCS is large enough to produce accurate weights for Ukraine’s eight economic zones plus Kyiv, but not for all 27 regions; current CPI uses same zone expenditure weights for regions within a zone while region-specific price samples exist.
  - Consideration: nine-zone level CPIs may be almost as useful as 27 regional CPIs, allowing smaller CPI sample and possible redeployment of regional staff to implement disaggregation methodology.

- Source data alignment and coverage:
  - HLCS is representative of the household population not institutionalized; expenditures collected at purchasers’ prices and aligned with recall period.
  - CPI price sample represents monetary transactions in urban areas for goods and services used by households, excluding used goods.

- Timeliness:
  - HLCS is continuous, producing quarterly household expenditure information with an eight-month processing lag permitting annual CPI weight updates for the previous year.
  - CPI sample processing lag allows index release within 10 days of the end of the reference month, exceeding SDDS requirements.

### 3.2 Statistical techniques
- Compilation techniques:
  - CPI uses sound statistical techniques.
  - Goods and services detail: 270 items.
  - Weights compiled on an accrual basis, accounting for payment arrears and including full purchase price of items on credit.

- Index calculation stages:
  - Stage 1 (regional offices): aggregate elementary item prices into a one-month change index for each of the 270 item groups in each of the 27 regions using a ratio of averages (Dutot formula).
  - Stage 2 (Price Statistics Department at headquarters): aggregate the 270 short-term regional indices into the national index using a modified Laspeyres formula.

- Identified methodological issues and recommendations:
  - Improvement needed in methodology of calculating item indices; SSCU aware and judges quantitative impact likely small because affected item groups (mainly fresh fruits and vegetables) have small expenditure weights.
  - Current practice does not carry forward imputed prices for temporarily missing elementary items; month-to-month chaining is used, which imputes short-term price change but is not “self-correcting” for long-term levels.
  - Recommendation: compute elementary item indices by updating imputed prices each month so they are “self-correcting” to ensure most accurate long-term item index levels.
    - Near-term implementation would require training 540 field staff and assuring revised calculation procedures are implemented.
    - Alternatively, centralize compilation in headquarters using a unified CPI database (similar to PPI practice) to handle imputation and index compilation by computer.

- Quality adjustment and new products:
  - Quality-adjusted direct comparisons are not performed when substitution or specification changes occur; new item is included after two months of data, which is equivalent to an implicit quality adjustment if overlap exists.
  - New products are detected via normal sample replacement; no special scanning for new products beyond replacement of permanently unavailable items.

- Coverage in relation to national accounts:
  - CPI expenditure aggregate covers 97 percent of national accounts household consumption.
  - CPI is the basis for the deflator for household consumption in the national accounts.
  - Household capital formation in the form of net acquisitions of dwellings is not covered in the CPI.

*Source: _cr03256 - 1. Integrity*

### 3.3 Assessment and validation of source data

### 3.3 Assessment and validation of source data

### Source data assessment practices
- Source data include censuses, sample surveys, and administrative records and are routinely assessed for coverage, sample error, response error, and nonsampling error; results are monitored and made available to guide planning.
- CPI weights are estimated for each economic zone from the HLCS; the HLCS sample design and methodology permit calculation of standard errors on expenditure estimates and of the impact of the nonresponse adjustments included in the household survey sampling weights.
- The Department of Household Surveys routinely assesses the accuracy of the standard aggregates produced from the HLCS.
- CPI prices: no sampling errors are available because the CPI does not use a random-sampling design.
- Unusually large movements in item indices are traced back to the reporting retail outlet if necessary and verified; verified prices and elementary indices are included in the index on verification or corrected if found to be in error.
- There are no procedures for eliminating unusually large movements in prices from the index.
- Administrative sources used in the CPI (e.g., rates for communal utility services, municipal transit fares) are presumed accurate and are not routinely checked except on the occasion of unusually large price movements.
- Attempts are made to keep the sample size large enough in covered urban areas so that prices are obtained for all 270 item groups in the CPI.
- All CPI source data for expenditure weights and prices are consistent with CPI concepts regarding price and expenditure definition, valuation, reference periods, and classifications.
- When CPI weights are updated, they are reviewed in terms of change in the weights and their differential impact on the index compared with the old weights.
- There are no formal and routine assessments of sample error or response and other nonsampling error, but attention is given to turning around nonrespondents or slow respondents, and this knowledge is used to guide survey planning.

### 3.4 Assessment and validation of intermediate data and statistical outputs

### Validation and discrepancy investigation
- Item indices from the CPI are routinely compared with similar product or industry indices from the PPI in SSCU’s annual publications.
- Unusual movements in item indices are checked and traced back to the reporting outlet if necessary.
- The CPI is imputed in only one way, so aggregating first by areas and then by items yields the same result as aggregating first by items and then by areas; therefore, there are no discrepancies arising from different imputation schema.
- Index weights are monitored annually as the index is revised and the sample of outlets reallocated toward items whose share has significantly increased.

### 3.5 Revision studies

### Revision practices and use
- The Ukrainian CPI is not revised once published; published data are final.
- The impacts of weight updates are examined to inform development of price samples.

### 4 Serviceability

### 4.1 Relevance
- The CPI and PPI enjoy an active user base; the Price Statistics Department receives a large number of contacts from users in and out of government, with government users a particular source of inquiries.
- The Department incorporates changes in its publications within resource and confidentiality constraints and makes special tabulations in some cases.
- A recent Presidential Order on price statistics is aimed at facilitating collaboration with users, but there is no formal advisory group or users’ survey to assess the relevance of the price programs.

### 4.2 Timeliness and periodicity
- The CPI and PPI are released within 10 days of the end of the reference month, bettering the SDDS requirement that they be published within one month of the reference month.
- The CPI and PPI are monthly indices, as required by the SDDS.

### 4.3 Consistency
- The CPI and PPI are imputed in only one way, so the order of aggregation does not affect the all items index; therefore, no discrepancies arise from different imputation schema.
- The CPI and PPI are annually chained.
- Disaggregated CPI data for the months of August through July, with base of the year prior to August equal to 100, may be aggregated using the previous year weights to obtain higher-level aggregates and the all items index.
- Owing to chaining, these aggregations are not exactly comparable between months with weights from different base periods.
- The imputation of missing prices arising from seasonal availability and supply interruptions is not “self-correcting” because imputations are not carried forward on the monthly pricing forms by the regional office staff.
- Items affected by this now comprise about 1 percent of index weight; the problem can be expected to become more serious as economic conditions improve.
- Item indices from the CPI are routinely compared with similar product or industry indices from the PPI; the oil and gas index for the PPI is compared with oil and gas data of a major public oil and gas corporation.

### 4.4 Revision policy and practice
- Index weights of both the CPI and PPI are updated annually with full year expenditure and output data, respectively, for the previous calendar year.
- New weights are normally introduced with data published in the June Express Report each year for the PPI and the August Express Report for the CPI.
- There are no revisions to published CPI or PPI data; published data are final.
- Studies of the impacts of annual weight revisions are not disseminated in existing publications but would be appropriate for the annual CPI and PPI bulletins.

### 5 Accessibility

### 5.1 Data accessibility
- A CPI index with December 1997 = 100 is published in the Government Courier every month as required by a 1998 law on indexation of wage arrears; however, there is no table containing a time series of all of these monthly CPI values calculated with a given reference period, so constructing such a table requires referencing many monthly issues of the Courier.
- The CPI and PPI annual bulletins contain a table showing the December values of these indices relative to December 1990, but there is no table showing all months over the published period relative to December 1990 or some other given reference period.
- The lack of a long time series of five years or greater length, with a given reference base, does not meet international good practice for index publication.
- Other analytical tables present changes since December of the previous year, since the same month of the previous year, and since the previous month, plus breakdowns by aggregates of goods and services; these could be generated from a single long time-series table.
- Data are transmitted simultaneously by mail, electronic mail, and fax to all interested parties.
- An advance release calendar is available on the SSCU’s website.
- Nonpublished special tabulations that satisfy SSCU’s disclosure requirements are available on request.

### 5.2 Metadata accessibility
- Methodology pamphlets on the CPI and PPI methodology have been produced and are in limited circulation within the government.
- A full methodology publication on major SSCU’s data series is in preparation for general release.
- SSCU’s publications and on-request tabulations meet a wide range of user preferences regarding display of detail by region and by product for the CPI and by region, industry, and product for the PPI.

### 5.3 Assistance to users
- A contact person with telephone number is provided on the inside of every price statistics publication.
- SSCU publishes a comprehensive print catalogue of all of its publications.

*Source: _cr03256 - 3.3 Assessment and validation of source data*

### 0.2 Resources

### 0.2 Resources

### Staff, financial, and computing resources (0.2.1)
- Overall numbers of staff appear adequate.
- Staff resources for the combined CPI, PPI, and construction price index programs comprising the SSCU’s Price Department were budgeted as follows:
  - State Statistics Committee (SSCU) — Total: 570; Consumer Price Index (CPI): NA; Production Price Index (PPI): NA; Construction Price Index: NA; Retail Price Index & Communication & Transport Price Indices: NA; Administration: NA; All price program positions: NA
  - Headquarters (Price Statistics Department) — Total: 21; CPI: 6; PPI: 5; Construction Price Index: 2; Retail Price Index & Communication & Transport Price Indices: 6; Administration: 2
  - Computer Center — Total: 9; CPI: ---; PPI: 6; Construction Price Index: 1; Retail Price Index & Communication & Transport Price Indices: 2; Administration: ---
  - Regions, of which — Total: 540; CPI: NA; PPI: NA; Construction Price Index: NA; Retail Price Index & Communication & Transport Price Indices: NA; Administration: NA; All price program positions: NA
  - Centers (regions) — Total: 308; CPI: NA; PPI: NA; Construction Price Index: NA; Retail Price Index & Communication & Transport Price Indices: NA; Administration: NA
  - Districts (regions) — Total: 232; CPI: NA; PPI: NA; Construction Price Index: NA; Retail Price Index & Communication & Transport Price Indices: NA; Administration: NA
- The current budgeting system tracks staff allocation by individual price program in central office headquarters and Computer Center units, but not in the regions, where individual staff may work on two or more programs.
- Financial resources:
  - Have improved recently with economic conditions.
  - SSCU resources come from the government budget and income from “subordinated bodies” (special tabulations, publication sales, computer and research services, specialized training such as accounting).
- Computing resources:
  - Headquarters: adequate; local area network of Pentium personal computers with standard word processing and spreadsheet software; used to compile the CPI from 270 item indices transmitted from each of the 27 regions each month and to prepare tables and charts for CPI, PPI, construction, and retail/communications/transport indices.
  - Budget is very limited for normal hardware and software upgrades.
  - Maintenance of the headquarters’ network is the responsibility of the SSCU’s Department of Computation (distinct from the Computer Center), which handles large-scale data processing.
  - Data receipt: CPI and PPI data from regions are transmitted via electronic mail to the Computer Center in Kyiv; maintenance of PPI microdata and central PPI calculation are undertaken at the Computer Center.
  - Regional computing: barely adequate. CPI short-term item indices are calculated in regions on computers or calculators and forwarded to headquarters. PPI data entry in regions uses old systems with limited memory, storage, and processor power, employing a custom-programmed Clipper database application. PPI calculation in the Computer Center uses another custom Clipper program.

### Measures to ensure efficient use of resources (0.2.2)
- SSCU’s Department of Finance and Department of Computation maintain staff utilization data for each SSCU department at headquarters, in the computer center, and in the regions; Price Statistics Department managers do not routinely use these data because regional utilization detail by program is lacking.
- Regional workload by program is monitored informally via communication with regional office managers.
- Staff compensation structure:
  - Comprises base salary, a performance based bonus salary, and cash awards for meritorious service.
  - SSCU budgets base and bonus salary by Department.
  - Bonus level for individuals determined by total annual bonus budget allocated to the Department and level of staff performance.
  - Awards for outstanding individual performance can be made over and above the Departmental compensation budget with adequate justification.
- Staff performance appraisal:
  - Follows civil service law and regulations; appraisals are given at three-year intervals.
  - The infrequency of appraisals could delay recognition or remediation of performance issues.
  - Flexibility: Departmental managers can review and set the bonus portion of staff salaries based on their judgment; normally done once a year but adjustments can be made at any time.

### Quality awareness (0.3)
- 0.3.1 Processes to focus on quality:
  - No formal quality program exists, but management is sensitive to quality issues and encourages adoption of procedures for quality checks during data collection, processing, and compilation.
  - Ongoing quality improvement projects are part of an inter-agency statistical improvement program: Statistical Reform Program (1998–2002) and Statistics Development Program (2003–2007).
  - CPI improvement project (2000–01): streamlined the CPI sample from 475,000 price records per month to 275,000 without harming national sample coverage; implemented COICOP/HBS classification.
  - PPI: implementing Eurostat’s NACE industry classification (derived from ISIC 3) and CPA product classification during the first half of 2002; PPI now compiled with an improved central database application.
  - Data models for CPI and PPI implement consistency with national accounts by design (common household expenditure survey for CPI weights and national accounts; industrial survey for PPI output weights and national accounts).
  - Time-series consistency between comparable components of CPI and PPI is examined monthly.
- 0.3.2 Processes to monitor quality of collection, processing, and dissemination:
  - Report forms reviewed by management and Methodology Division, cooperating with National Accounts Department.
  - Methodology Division tests report forms with a small group of enterprises.
  - Input-output compilation and reconciliation serve as checks on system of national accounts components.
  - Reviews undertaken of all unusual movements in price data at regional and central levels and traced back to reporters if necessary; for the PPI, personal visits occur twice a year to verify responses received principally by mail.
  - Regional office error performance in data collection and elementary index calculations is routinely tracked; problems addressed with regional managers.
  - SSCU’s General Information Department checks all publications for accuracy; quarterly and annual reviews of regional publications are undertaken with recommendations on accuracy, format, presentation, quantitative and descriptive content.
  - No body distinct from SSCU provides guidance on series quality and strategies for improving data production; revival of a high-powered Statistical Council attached to the office of the President is noted as a possibility.
  - User feedback: under Presidential Regulation on the Observation of Prices and Tariffs (2001–03), government policy users were surveyed; no formal surveys of nongovernment users. Price Department receives routine user contacts:
    - 40-50 letters a month on price programs as a whole; 543 letters on the CPI alone during 2001; high volumes of electronic mail via the SSCU website and frequent telephone inquiries.
- 0.3.3 Processes to handle quality trade-offs and guide planning:
  - Price indices are required by regulation to be released by the tenth of the month following the reference month; Price Statistics Department staff report no conflict between timeliness and quality.
  - All checks are routinely completed within the available processing and review window, allowing some flexibility for regional reporting.

### Integrity and transparency (summary of sections 1.1–1.3)
- Professionalism (1.1):
  - Article 5 of the Statistics Law ensures professional independence of SSCU and forbids interference by state/local authorities, officials, public associations, and other persons.
  - SSCU is an independent organization with its own financial accounts and identifiable seal; chairperson appointed by the President on Prime Minister’s recommendation (Presidential Decree of November 6, 1997 (No. 1249/97)).
  - Civil service law determines staff recruitment qualifications; professional skills in statistics and economics are considered in compensation and promotion.
  - Articles 5, 7, 8, and 9 of the Statistics Law guarantee independence in methodology, data sources, content, and format; Price Department reports no external interference.
  - Article 13 empowers SSCU to comment on misuse or misinterpretation of statistical information; SSCU communicates disagreements in print, other public modalities, or direct private communication.
- Transparency (1.2):
  - Statistics Law is public and available on SSCU official website; statistical publications provide contact information.
  - Article 14 guarantees equal access and simultaneous release to all interested parties; no access prior to release is provided to anyone outside SSCU.
  - Government publications must attribute SSCU as the source.
  - Advance notice of methodological changes is provided in programs of statistical development and the annual SSCU plan (public and provided to major users); methodological changes are noted in CPI methodological manual and annual bulletin after implementation. Recommendation: an additional alert in the monthly Express Report prior to implementation is advisable.
- Ethical standards (1.3):
  - Article 17 of the Statistics Law requires employees to protect confidential information; Civil Service Code provides penalties for misuse; staff informed and must commit to these regulations as a condition of employment.

### Methodological soundness (sections 2.1–2.4)
- Concepts and definitions (2.1):
  - PPI concepts and definitions fit within the requirements of the System of National Accounts 1993 (1993 SNA) and broadly follow internationally accepted good practices.
- Scope (2.2):
  - PPI covers mining, manufacturing, energy generation and distribution, and other utilities such as municipal steam generation (NACE groups C–E).
  - PPI does not cover shipbuilding, aircraft production, and military production.
  - Coverage of more market-service activities would be desirable.
  - Output weights cover a full year and are compiled from a monthly industry survey by the Industrial Statistics Department.
  - Enterprises in free trade zones are covered.
- Classification/sectorization (2.3):
  - Classification of transactions and institutional units follows the 1993 SNA.
  - Classification of establishments by activity follows Eurostat’s NACE, Revision 1.
  - Classification of products follows Eurostat’s CPA.
- Basis for recording (2.4):
  - Market prices are used to value flows and stocks: output produced in a given month is valued at prices at which units are sold in that month by the establishment.
  - Recording is done on an accrual basis: output booked in the month in which it is produced (1993 SNA accrual principle).
  - Grossing/netting: Ukrainian PPI is weighted on a gross output basis; no stage of processing indices produced, so no grossing/netting on that score.

### Accuracy and reliability (sections 3.1)
- Source data collection (3.1.1–3.1.3):
  - Monthly industry survey used in the PPI comprises a 50 percent cutoff sample by four-digit NACE industry plus a judgmental sample of small- and medium-sized establishments; this covers 80 percent of the aggregate output of the covered industries.
  - Estimates produced for four- or five-digit NACE, and five-digit CPA strata; elementary item level weights at a nine-digit level (five-digit CPA plus four digits) to identify product varieties.
  - Regional PPI staff continuously consult region industry statistics staff to pick up new product varieties; new goods and services picked up during annual updates of output weights and during replacement of elementary items.
  - Source data reasonably approximate required definitions, scope, classification, valuation, and time of recording.
  - Timeliness:
    - Weights are sufficiently timely to permit annual updates of the weights with previous year output data by June.
    - Price data are sufficiently timely to release the index within 10 days after the end of the reference month, exceeding SDDS timeliness requirements.

*Source: _cr03256 - 0.2 Resources*

### 3.2 Statistical techniques

### _cr03256 - 3.2 Statistical techniques

### 3.2 Statistical techniques and procedures
- Index formula: Laspeyres at all levels of aggregation.
- Treatment of missing sales: "If there are no sales in a given month, the prices prevailing in the most recent month for which there were sales are used."
- Imputation approach: Imputations for missing prices are done only one way—by product—for both product and industry price indices, ensuring consistency between breakdowns and the all items index.
- Other statistical procedures (data adjustments, transformations, statistical analysis): Not applicable.

### 3.3 Assessment and validation of source data
- Weights source:
  - Source data for the weights are routinely assessed and made available by the Industry Statistics Department to the Price Statistics Department to guide PPI compilation.
- Price data:
  - Source data for prices are routinely assessed by the regional offices and the results made available to the Price Statistics Department.
- Error assessment:
  - "There are no formal and routine assessments of sample error or response and other nonsampling error," but attention is given to turning around nonrespondents or slow respondents and this knowledge is used to guide survey planning.

### 3.4 Assessment and validation of intermediate data and statistical outputs
- Cross-validation:
  - "The PPI is compared with the CPI for comparable products between the CPA and COICOP classifications."
  - Item indices from the CPI are routinely compared with similar product or industry indices from the PPI in SSCU’s annual publications.
  - "The oil and gas index for the PPI is compared with the oil and gas data of a major public oil and gas corporation."
- Investigation of discrepancies:
  - "Unusual movements in the index arising from large movements in particular sectors or from particular reporters are investigated and errors corrected."
  - Explanatory tables are provided in the annual bulletin to identify sources of index change.
- Imputation consequence:
  - Single imputation method by product ensures product and industry breakdowns produce the same all items index.

### 3.5 Revision studies
- Revision practice:
  - "The Ukrainian PPI is not revised once published. However, the impacts of annual weight updates are examined to inform development of price samples."
- Dissemination of revision analysis:
  - Studies of the impacts of annual weight revisions are not disseminated in existing publications but "would be appropriate for the annual CPI and PPI bulletins."

### 4. Serviceability — relevance, timeliness, periodicity, consistency, and revision policy
- Relevance and user engagement:
  - "The CPI and PPI enjoy an active user base" with many contacts from users in and out of government, especially government users.
  - No formal advisory group or users’ survey to assess relevance.
  - Special tabulations are provided within resource and confidentiality constraints.
- Timeliness:
  - "The CPI and PPI are released within 10 days of the end of the reference month, bettering the SDDS requirement that they be published within one month of the reference month."
- Periodicity:
  - "The CPI and PPI are monthly indices, as required by the SDDS."
- Consistency within dataset:
  - Single imputation method ensures order of aggregation does not affect the all items index.
- Consistency over time:
  - "The CPI and PPI are annually chained."
  - Disaggregated CPI data for August through July with base of the year prior to August equal to 100 may be aggregated using the previous year weights to obtain higher-level aggregates and the all items index; "owing to chaining, these aggregations do not obtain on any other time-series comparison."
- Consistency with other sources:
  - CPI item indices are compared with PPI indices; PPI oil and gas index compared with major public oil and gas corporation data.
- Revision policy and practice:
  - Index weights for both CPI and PPI are updated annually with full year expenditure and output data for the previous calendar year.
  - New weights are normally introduced with data published in the June Express Report each year for the PPI and the August Express Report for the CPI.
  - "There are no revisions to published CPI or PPI data; published data are final."

### 5. Accessibility — data and metadata accessibility, assistance to users
- Data presentation:
  - No single long, multi-year time series table of monthly PPI values with a given reference period exists; users must reference many monthly issues to construct such a table.
  - Annual bulletins contain a table showing December values relative to December 1990, but there is no table showing all months over the published period relative to December 1990 or another given reference period.
  - Other analytical tables show changes since December of the previous year, since the same month of the previous year, and since the previous month, with breakdowns by aggregates.
- Dissemination media and simultaneity:
  - Data are transmitted simultaneously by mail, electronic mail, and fax to all interested parties.
  - "An Advance Release Calendar is available on the SSCU website."
- Nonpublished aggregates:
  - Nonpublished special tabulations that satisfy SSCU’s disclosure requirements are available on request.
- Metadata:
  - Methodology pamphlets on the CPI and PPI have been produced and are in limited circulation within the government.
  - "A full methodology publication on major SSCU’s data series is in preparation for general release."
  - Documentation is available but "is not timely and could be more comprehensive."
- User assistance:
  - A contact person with telephone number is provided on the inside of every price statistics publication.
  - SSCU publishes a comprehensive print catalog of all of its publications.

### Key numeric and procedural facts (verbatim)
- Index formula: Laspeyres.
- Release timeliness: within 10 days of the end of the reference month.
- SDDS timeliness requirement: "within one month of the reference month."
- Periodicity: monthly.
- Annual weight updates: introduced in the June Express Report for the PPI and the August Express Report for the CPI.
- Reference-base table available: December values relative to December 1990.
- Imputation: done only one way—by product—for both product and industry price indices.
- Revision practice: "The Ukrainian PPI is not revised once published."

*Source: _cr03256 - 3.2 Statistical techniques.*

### 1. Integrity

### 1. Integrity

### Professionalism
- Statistics are compiled on an impartial basis.
  - Article 8 of the Budget Code requires the MoF to authorize budget classifications; the minister acts on the advice of his professional statisticians.
  - Article 58 gives responsibility for designing reports on Budget execution to the STU in coordination with the MoF and the Accounting Chamber of Ukraine.
  - A working group of senior members of these three agencies, chaired by the State Secretary of the MoF, supervises the compilation of fiscal statistics.
- Choices of sources and statistical techniques are informed solely by statistical considerations.
  - Staff of the MoF and the STU are free from political or other influence in choosing sources and methods for compiling GFS.
- The appropriate statistical entity is entitled to comment on erroneous interpretation and misuse of statistics.
  - The Head of the Financial and Economic Statistics Division can (and frequently does) respond publicly to erroneous interpretation of fiscal statistics.
  - Explanatory material is provided to the mass media on request.

### Transparency
- Terms and conditions for collection, processing, and dissemination are published in the Budget Code and the Budget Law, and in the Government Courier (Uryadovy Courier).
- No government ministers or officials have access to the data before their release.
- MoF publications are clearly labeled as statistical publications; individuals involved are identified on the inside cover. Tables in the SSCU’s publications are attributed to the STU.
- Advance notice of methodological changes:
  - The annual Budget Law specifies methodology for the forthcoming year; consequent MoF orders elaborate the differences from the previous Law.
  - Such notices are aimed primarily at budget management and are not convenient to users of statistics.

### Ethical standards
- Staff are bound by general rules in the Civil Service Code.
- There are no special guidelines for behavior of staff compiling and disseminating GFS.

### Methodological soundness

- Concepts and definitions
  - Fiscal statistics generally follow A Manual on Government Finance Statistics, 1986 (GFSM 1986).
  - Movement toward GFSM 2001: adoption of the revised COFOG and treating privatization receipts as financing rather than revenue.
  - Plans for migration to GFSM 2001 are part of State Statistical System reform; timing of later stages depends on the provision of aid.

- Scope
  - Monthly data published relate to budgetary central government; quarterly statistics relate to budgetary general government.
  - Published data omit transactions of the nonbudget pension fund and social security fund; information on these funds is available quarterly but not included in published central or general government data.
  - Published time series reflect changes in coverage.
  - Table (as presented):
    - Revenue                                  Expenditure
    - 2001
    - Billion Hryvnia    Percent of General Government    Billion Hryvnia    Percent of General Government
    - Pension and Social Security Funds* 20.8 28.5 19.6  26.1
    - Summary General government 73.1  75.0
    - Source: Ministry of Finance
    - *Includes Pension, Social Security, Unemployment, and Accident Funds
  - Development plan includes compilation and publication of statistics on the full international definition of central, local, and general government.

- Classification/sectorization
  - Economic classification of transactions and classification of outstanding debt are consistent with GFSM 1986.
  - Functional classification from 2002 is consistent with GFSM 2001 (1998–2001: GFSM 1986; previously a national functional classification).
  - Revised COFOG has yet to be extended to local government.
  - Payments fulfilling government guarantees on debt are treated as payments of debt charges.

- Basis for recording
  - Market prices: Cash transactions reflecting actual prices are recorded in GFS.
  - Recording basis: The budgetary system operates on a cash basis; expenditure recorded when checks clear and revenue when checks are paid into the bank.
  - Since 1998, data on arrears (amounts receivable and payable) classified by economic category and function have been collected to produce estimates of accruals.
  - Grossing/netting: Practice aligns with international recommendations; enforced by Budget Code Article 56(4).

### Accuracy and Reliability

- Source data
  - Sources are government accounting data for government operations and the debt database of the State Debt Department (SDD).
  - Budget Code (Article 58) authorizes the STU to obtain reports from all spending units, including extra-budgetary funds; STU compiles detailed summary reports for MoF.
  - Local government sector is covered by the STU budget management system.
  - From 2002, accounts are consolidated at the relevant level; previously coding of transfers did not distinguish level of government.
  - From July 1, 2002, all central government budgetary spending units should be covered by TSA.
  - From January 1, 2004, local government expenditures should be covered under the TSA; local government revenues are already handled by the Treasury.
  - From 2003, after planned implementation of the new centralized ledger accounting system, expenditure data would be aggregated automatically to provide GFS; revenue statistics will continue to be calculated by bridge tables in MSExcel spreadsheets.
  - Debt database maintains outstanding debt at nominal values and in original currencies; revaluation to hryvnia is done outside the database.

- Source data timeliness
  - Budget Code (Articles 58–61) specify timing of reports; timing is consistent with SDDS standards and is met in practice.

- Statistical techniques and validation
  - Compilation is by aggregating accounting data; no special statistical techniques required.
  - Collection of cumulative year-to-date cash data leads to extremely small revisions: the revisions between the first 12-month estimates and final audited annual data are of the order of 0.1 percent; these revisions are attributed to the final month.
  - MoF does not compare or reconcile its data with other information (accounting information provides full coverage).
  - The STU reconciles treasury ledger accounts with NBU bank accounts on a regular basis, but aggregate data on net credit to government are not reconciled with monetary survey data.
  - Any discrepancy in balancing the fiscal accounts would be investigated and resolved.
  - Revision studies: Revisions are minuscule.

### Serviceability

- Relevance
  - No formal surveys of user requirements; staff are frequently in contact with users for data requests.
  - Budget sector statistics, using GFS-consistent classifications, are used within government to monitor budget sector impact and for fiscal analysis.
  - GFS compiled in 2001 for GFSM 1986 were prepared solely for publication in the IMF’s Government Finance Statistics Yearbook and are not used or published in Ukraine.
  - Users are largely unaware of potential uses of GFS; little demand outside government, reflecting lack of complete and easily understandable GFS.

- Timeliness and periodicity
  - Timeliness largely meets SDDS standards:
    - Monthly central government operations statistics are available approximately 25 days after the end of the reference period but released to the general public after a lag of 35 days.
    - Quarterly debt figures are published within 45 days.
    - Annual operations data based on audited accounts are required by the Budget Code to be published by June 1, five months after the end of the year.
  - Periodicity meets SDDS standards:
    - Central government operations data: monthly and annually (subject to scope qualifications).
    - General government operations data: quarterly.
    - Central government debt statistics: monthly, quarterly, and annually.

- Consistency
  - Statistics are consistent within the dataset due to the derivation system.
  - Statistics are consistent or reconcilable over time only to the extent that published data relate to the budget coverage in that year; differences in scope or classification between years are explained in the annual publication and in Budget Laws and MoF Orders, but breaks in series are not indicated on tables.
  - No reconciliation of treasury data on net credit to government with information from the monetary survey.

- Revision policy and practice
  - Revisions arise only from incorporation of audited annual accounts, occurring in June each year.
  - The first estimate for the year—the 12-month estimate—is labeled as preliminary.
  - Studies and analyses of revisions are made public; revisions are described as minuscule.

### Accessibility

- Data accessibility and formats
  - GFS are available monthly but do not provide the equivalent coverage set out in GFSM 1986; data formats suit short-term budget execution monitoring rather than broader fiscal analysis.
  - STU maintains databases of input and summary data; a separate database is opened for every year reflecting that year’s Budget Law. Databases are accessible to STU headquarters’ staff; detailed data are sent electronically to MoF, SSCU, and the Ministry of Economy; earlier years are archived.
  - MoF is considering creation of a database accessible to outside users as part of development plans.

- Dissemination media and scheduling
  - Publications containing fiscal statistics (frequency and publisher listed in source).
  - SSCU has a website where fiscal data are irregularly posted; MoF has recently set up a website that publishes fiscal data but with a two-day delay after release before appearing on the site.
  - An advance release calendar (ARC) is disseminated on the SSCU website (www.ukrstat.gov.ua) but is not maintained current; availability of the ARC will be publicized in the future in the MoF’s Monitoring of Budget Execution.
  - Government operations data and debt data are released simultaneously to all interested parties via flash estimates and publications; quarterly debt bulletin copies can be obtained on request from the MoF and are posted on the MoF’s website.
  - Unpublished (but nonconfidential) sub-aggregates are made available upon request, but terms and availability are not publicized.

- Metadata and user assistance
  - Documentation exists in the form of the Budget Code and Laws; other than summary notes in some periodicals, there is no documentation aimed specifically at users and laws do not make specific reference to international standards or the accounting framework.
  - Compilers in the MoF are named on the inside cover of MoF’s statistical publications; a general MoF telephone number and address are provided.
  - The MoF produces no catalogues of publications, documents, or services.

*Source: _cr03256 - 1. Integrity*

### 0. Prerequisites of quality

### 0. Prerequisites of quality

### Legal and institutional environment
- The National Bank of Ukraine (NBU) is the sole official agency in charge of collecting, compiling, and disseminating monetary statistics.
- Legal basis: Law of Ukraine on the National Bank of Ukraine, No. 679–XIV of May 1999, as amended on January 10, 2002.
  - Article 57: NBU entitled to obtain necessary information from government agencies and businesses for banking and financial statistics and economic analysis.
  - Article 67: NBU authority to request any information from banks and financial institutions located on the territory of Ukraine and to define reporting forms and procedures.
  - Article 68: NBU should issue a monthly statistical bulletin (Visnyk of the NBU), publish current banking information on monetary statistics (subject to state and banking secrets), and provide monetary data in accordance with international treaties.
  - Article 53: NBU shall enjoy absolute autonomy in exercising powers and duties under the law.
  - Article 69 (Law on Banks and Banking Activity): banks required to submit annual balance sheets and income statements, reports of boards of directors and auditors, and reports of independent auditing institutions after the date of preparation.
- Resolution on confidential information: Board of the NBU resolution on The Procedure of Handling Confidential Information in the NBU adopted on December 12, 2001; approved regulation and instructions for protection and handling of confidential and “not for publication” information.
- Data sharing and coordination: arrangements within NBU ensure flow of information among Accounting Department, Information Technologies Department, Economic Department’s Monetary and Banking Statistics Division, Monetary Policy Department, Foreign Exchange Regulation Department; MoF, SSCU, and STU involved in SDDS preparations with inter-agency meetings when necessary.
- Confidentiality: Articles 57 and 67 forbid publication/disclosure of private and personal statistical information except as legislated; access limited to authorized staff; computers password protected; consolidation/aggregation procedures prevent individual disclosure.

### Resources
- Monetary and Banking Statistics Division staffing: total 24 staff, 11 of which are professionals primarily devoted to the compilation of monetary statistics; 5 of them are also responsible for statistical concepts and methodology, and improvement of coverage and accuracy.
- Training and professional development: staff participate in IMF training, seminars of other central banks, internal NBU training; employment competitive; staff turnover not high.
- Computerization: data collection system highly computerized; reporting financial institutions submit returns through Information Technologies Department which performs validation checks; NBU’s and commercial banks’ balance sheets available daily.
- Budgetary constraints: currently no budgetary constraints impeding data collection and compilation activities; possible need to expand human and budget resources if institutional coverage of monetary data and presentation of flow of funds for the financial sector are expanded per MFSM recommendations.
- Enforcement: Resolution of the Board of the NBU of May 26, 2000 No. 215 on enforcement measures against banks/financial institutions for violating reporting requirements; fines in case of incorrect, missing, or late submission: 500 times the nontaxable minimum monthly income. In practice, penalties have not been enforced; NBU considers nonresponse not possible because all banks must be connected to automated processing system.

### Quality awareness
- Quality control: exercised at every stage of data production and dissemination; cross-checks of source data; discrepancies investigated; validation procedures include visual bank-by-bank plausibility checks.
- Monitoring and user contact: Monetary and Banking Statistics Division responds monthly to about 60 questions related to monetary data; no formal user surveys conducted; users can contact Economic Department via website.
- Internal quality roles: specific staff assigned to monitor financial market developments and improve methodological soundness; undertaking verification and analysis of investment and private pension funds for possible inclusion in monetary statistics based on IMF technical assistance recommendations.
- Trade-offs: recognition of trade-offs between data quality and timeliness; production fully automated with checks and validations at every stage; timeliness regarded as critical in line with SDDS requirements.
- Suggested improvement: establishment of periodic surveys of users of the monthly Bulletin of the NBU to request reactions on usefulness and needs for specific detail.

### 1. Integrity

### Professionalism
- Compilation impartiality: statutory provisions support independence for compilation of monetary statistics.
- Professional development: staff encouraged to participate in lectures, conferences, meetings with other compilers and international organizations.
- Source and technique choices: informed solely by statistical considerations.
- Commentary on misuse: Economic Department comments on erroneous interpretation if deemed significant; explanatory notes provided in publications; NBU management holds press conferences in cases of significant misinterpretations.

### Transparency
- Availability of terms/conditions: NBU Law is available on NBU’s website.
- Advance release calendars (ARCs) for SDDS: up-to-date ARCs for Ukraine’s SDDS data categories are currently not available to the public; drafts for SDDS financial sector data categories provided to SSCU and posted on SSCU’s website but do not meet SDDS requirements because no-later-than dates are not updated the week prior to release to identify a precise day.
- Internal governmental access: data released to the Administration of Ukraine’s President, Cabinet of Ministers, Ministry of Economy, and SSCU within about 20 days after the end of the reference month (about ten days prior to public release) with indication they can be used for internal analysis and cannot be published before official NBU dissemination; there is not public awareness of such internal government access prior to release.
- Product identification and attribution: Economic Department of the NBU identified as source of monetary statistics in the monthly Bulletin of the NBU; the Bulletin indicates it should be used as a source when reproduced, but NBU does not directly request attribution.
- Advance notice of methodological changes: advance notice given for major changes (e.g., introduction of new charts of accounts); other changes in classification, source data, and techniques are announced simultaneously with data release; explanatory notes provided when interpretation of time series changes is needed.

### 2. Serviceability, Methodological and Access Observations (selected points from Government Finance Statistics summary)
- Resources: "More resources will be needed to carry out existing development plans. These aim to migrate to the GFSM 2001."
- Transparency in methodology notice: "The Budget Laws and MoF Orders give advance notice of changes in methodology, but these are aimed primarily at budget management and are not convenient to users of the statistics."
- Scope limitations: "Published statistics exclude pension and social security funds."
- Classification gaps: "The COFOG functional classification has yet to be extended to local government."
- Revisions: "Revisions are negligible."
- User relevance and monitoring: "No formal monitoring of user requirements, although there is frequent contact with users in meeting data requests."
- Timeliness: "Central government operations are not disseminated within prescribed SDDS timeliness."
- Consistency: "No backward revisions as coverage has increased over recent years as additional formerly extra-budgetary funds have been brought into the scope of the budget."
- Release calendar and metadata: "An advance release calendar is disseminated on the website of SSCU but is not maintained current." "Documentation is contained in Budget laws but only briefly with the data." "Contact names identified in publications but no catalogue is issued."

*Source: _cr03256 - 0. Prerequisites of quality*

### 1.3 Ethical standards

### 1.3 Ethical standards

### Guidelines for staff behavior
- Guidelines are established in the NBU Law Articles 65 and 66 and in the Law on Banks and Banking Activity, Section 10, Articles 60, 61, and 62.
- Article 65 (NBU Law): officials of the NBU are prohibited from becoming members of managing bodies or shareholders of commercial banks and other financial institutions.
- Article 66 (NBU Law): officials of the NBU are prohibited from disclosing information that constitutes a service secret or is of confidential nature, and this prohibition continues even after resignation from the NBU, except for cases provided for by the laws of Ukraine.
- Articles 60–62 (Law on Banks and Banking Activity): provide details on information considered as banking secrets, actions needed for their protection and procedures, and cases for their possible disclosure.

### Methodological soundness (overview)
- The NBU’s framework for analytical accounts of the NBU (central bank) and commercial banks is based in general on the IMF’s Draft Guide to Money and Banking Statistics in International Financial Statistics (December 1984).
- Following the IMF’s MFSM publication in September 2000, the NBU intends to revise procedures and formats for collection, compilation, and dissemination of monetary statistics and has established a working commission to harmonize use of SNA and MFSM terminology in normative acts.
- The NBU is conducting studies to expand coverage of depository corporations to include nonbank financial institutions that may issue deposit-substitute instruments and to consider compiling a Financial Corporations Survey; expansion depends on appropriate legislation.

### Monetary aggregates and other data definitions
- Money aggregates compiled and disseminated by the NBU:
  - M0: currency in circulation.
  - M1: comprises and broken down into M0 and settlement and current accounts (enterprises, households, and nonbank financial institutions) in the national currency.
  - M2: comprises and broken down into M1 and time deposits (enterprises, households, and nonbank financial institutions) in the national currency, and all deposits (enterprises, households, and nonbank financial institutions) in foreign currencies.
  - M3: comprises and broken down into M2 and client funds arising from banks fiduciary transactions and securities issued that are held by other domestic sectors than commercial banks.
- Other aggregates: domestic credit disaggregated into credits to the general government (central government and local governments; extrabudgetary funds included), financial sector (excluding banks), private and public nonfinancial enterprises, and households; and components of the net foreign position published in a Ukrainian Page in International Financial Statistics in the monthly Bulletin.

### Scope and institutional coverage
- Depository corporations sector comprises the NBU and 154 active commercial banks (as of December 31, 2001).
- No branches of foreign banks operating in Ukraine (at that date).
- National definition of “bank” (Article 2, Law on Banks and Banking Activity): legal entity licensed by NBU with exclusive right to (i) accept deposits and funds of households and legal entities, (ii) allocate these funds in their own name at their own risk, and (iii) open and service accounts of individuals and legal entities.
- Institutional coverage includes banks’ domestic headquarters and domestic branches; branches of Ukrainian banks abroad are excluded.
- Monetary aggregates exclude data of 35 commercial banks that are in the process of liquidation; the Economic Department disseminates memorandum items on credits and deposits of nonoperational banks in the monthly Bulletin and on the NBU website, based on special reports from the Liquidation Bank Division of the General Banking Supervision Department.
- Other Depository Corporations (ODC) sub-sector (MFSM concept) would currently have the same coverage as the compiled analytical accounts because commercial banks are the only deposit-taking institutions meeting ODC criteria in Ukraine.
- Some financial corporations (investment funds, private pension funds, other corporations issuing securities sold to nonbank financial corporations) appear not to be included in the national broad money definition at present.
- Definition of residency follows BPM5 and 1993 SNA: branches of foreign banks in Ukraine are residents; branches of domestic banks abroad are nonresidents.

### Classification and valuation
- Residency criterion is used to distinguish domestic and external accounts; sectorization is principally in line with the MFSM.
- NBU’s detailed sectoral groupings: (i) central bank, (ii) commercial banks, (iii) nonbank financial institutions, (iv) general government, (v) nonfinancial public enterprises, (vi) nonfinancial private enterprises, (vii) households, (viii) nonprofit organizations serving households.
- Claims on and liabilities to central and local governments are grouped as general government; aggregate net claims on the general government are used rather than net claims on the central government.
- Financial derivatives data (other than foreign exchange swaps) are not available and not included in monetary statistics.
- Securities repurchase operations are treated as collateralized loans.

### Basis for recording and valuation practices
- Valuation principle: market prices or market price equivalents (fair values) are used; loans are valued at book value (outstanding principal plus accrued interest) without adjustment for expected loan losses.
- NBU and commercial bank accounting follow Ukrainian legislation, NBU Law, and International Accounting Standards; securities purchases recorded at nominal value with separate accounts for unamortized premiums and discounts to enable market-value calculation.
- NBU records foreign equity shares and gold at market value and revalues foreign equity at current exchange rates; commercial banks value gold at prevailing market prices on reporting date.
- Provisions for expected loan losses are recorded as separate liability entries and included in capital accounts in analytical accounts.
- All foreign-currency stocks and flows are converted to national currency at the market exchange rate prevailing at entry; midpoint between buying and selling rates is used; holding gains/losses from exchange-rate changes are recorded in a revaluation account.
- The NBU compiles transaction data for loans internally but does not disseminate these transaction breakdowns.

### Recording conventions, grossing and netting
- Recording on an accrual basis: interest due but not paid is incorporated into outstanding amounts; Ukrainian accounting requires interest to be accrued monthly; monetary data separately identify accrued interest and include it in the underlying instrument.
- Transactions are generally recorded at the time they occur via simultaneous electronic recording.
- Assets and liabilities of the NBU and commercial banks are collected and compiled on a gross basis; net presentations in the monetary survey are accompanied by underlying gross data.
- Monetary survey is obtained by consolidating and canceling out outstanding claims and liabilities between NBU and commercial banks and among commercial banks.

### Accuracy and reliability — source data and timeliness
- Source data: balance sheet records and supporting schedules from NBU Accounting Department and Information Technologies Department, and data from other NBU departments; these records capture the full range of financial instruments and sectors and permit classification per MFSM.
- Financial press is monitored for market developments; meetings with market participants and business community to identify new developments are not common.
- Positions not allocated to specific instruments or sectors are recorded in “other assets” and “other liabilities”; these amounts are small overall.
- Timeliness: accounting records are available daily and provided electronically to the Monetary and Financial Statistics Division; the final NBU monthly balance sheet is produced about 15 days after the reference month because data on NBU internal organizations (mint factory, training center) are provided with a delay of about two weeks after the reference month.
- The NBU indicates it could disseminate analytical accounts of the central bank excluding internal organizations within two weeks after the reference month using daily accounting records, in line with SDDS requirements.

### Statistical techniques and data processing
- Source files are provided in MSExcel format with built-in checks; compilation and processing are automated, using computerized files with macros to avoid processing errors; procedures are documented and supported by the Information Technologies Department.
- Full computerization of accounting procedures makes nonreporting of active banks’ data for monetary statistics practically impossible.
- Adjustments are made for positions with the IMF using the exchange rate at the end of the reporting period rather than the IMF Treasurer’s “holding exchange rate”.
- The Economic Department does not calculate seasonally adjusted monetary data but is examining Eurostat-recommended methods for seasonal adjustment.

*Source: _cr03256 - 1.3 Ethical standards (https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2003/_cr03256.pdf)*

### 3.3 Assessment and validation of source data

### 3.3 Assessment and validation of source data

### Assessment of source data and institutional handling
- Source data (censuses, sample surveys, administrative records) are routinely assessed for coverage, sample error, response error, and nonsampling error; results are monitored and made available to guide planning.
- Queries on balance sheet data and supporting schedules of the NBU and commercial banks are resolved between the Accounting Department, the Economic Department, and other NBU departments.
- Balance sheet records and supporting schedules are revised to reflect changes in the chart of accounts of the NBU and commercial banks and are rarely modified to cover sector reclassifications and compilers’ needs.
- The Economic Department is authorized to make revisions only in supporting schedules designed to collect statistical information for compilation of monetary statistics.
- If necessary, the Economic Department conducts meetings with the Accounting Department, which has authority to introduce changes to charts of accounts.
- A designated staff of the Monetary and Banking Statistics Division monitors developments and changes in financial markets that may affect compilation of monetary statistics.

### Validation against secondary sources and handling of discrepancies
- Accuracy of balance sheets and supporting-schedule data is checked against secondary sources where deemed necessary (e.g., data collected by other NBU departments), though most queries are resolved directly with banks by the Monetary and Banking Statistics Division.
- The Monetary and Banking Statistics Division investigates statistical discrepancies in intermediate data and determines major contributing factors.
- Large fluctuations in monetary or credit aggregates are rare and typically follow significant changes in methodology or source data; upon such releases the Division receives and handles inquiries directly.
- In cases of large unexplained fluctuations, the Division investigates possible classification/sectorization errors by individual respondents.

### Revision studies
- There is no mechanism for routine revision studies; NBU considers monetary data final when first published.
- Deviations, omissions, and other potential problems (e.g., erroneous sectorization of institutional units) are identified and investigated on an ad hoc basis.
- Results of these ad hoc studies are not made available to users.

---

### 4 Serviceability

### Relevance
- No formal process of regular consultation with policy departments, ministries, private sector, or academia; the Economic Department conducts informal reviews with NBU departments (notably Monetary Policy Department and Foreign Exchange Regulation Department) and with external users via e-mail, telephone, and facsimile.
- The Economic Department regularly participates in international statistical meetings and seminars.

### Timeliness and periodicity
- Analytical accounts of the central bank and the banking sector are available within about 20 days after the end of the reference month for IMF operational use and IFS publication.
- Both analytical accounts are disseminated with a lag of about 30 days after the end of the reference month when simultaneously released in the Bulletin of the NBU and on the Internet website.
- The NBU would be able to meet the prescribed SDDS criteria for timeliness of the analytical accounts of the central bank and disseminate on the Internet preliminary central bank aggregates (foreign positions, reserve money, and domestic credit) within two weeks after the end of the reference month; these aggregates would be based on daily accounting data available to Economic Department staff.
- Periodicity: analytical accounts of the central bank and banking sector are disseminated on a monthly basis, consistent with SDDS specifications.

### Consistency
- Central bank and commercial bank records for interbank claims and liabilities usually show only negligible discrepancies due to timing and accrued interest recording differences.
- No reconciliation of stock and flow data is available because flow data for monetary statistics are not compiled for all recommended components.
- NBU data on its claims on commercial banks show a discrepancy of 444 million hryvnia as of end-December 2001 when compared with commercial banks’ data on liabilities to the NBU, mainly due to classification of the NBU’s credit to a large bank that became inoperative (NBU classifies as claims on all commercial banks while data of the large inoperative bank are excluded from commercial banks’ data used in monetary statistics).
- Time series are available electronically and on the NBU website; when changes in source data, methodology, or techniques are introduced, all time series are revised and main breaks/discontinuities are explained in attached notes/footnotes.
- Consistency is hampered by very frequent changes to charts of accounts of the NBU and commercial banks and by exclusion of liabilities of banks in liquidation, creating breaks in monetary aggregates’ time series.
- Economic Department regularly checks consistency of monetary statistics with balance of payments statistics. Consistency checks between GFS and monetary statistics are not conducted on a regular basis.
- The NBU publishes monthly information on holdings of government securities bought by the NBU and commercial banks on the primary market to provide a partial indication of consistency between monetary statistics and GFS.

### Revision policy and practice
- No formal revisions policy; revisions are infrequent and made when more accurate data become available. Revised data are identified as such in publications. Users are not informed that NBU does not conduct routine data revisions.
- Monetary data are considered final when first released; Bulletin of the NBU does not include notes on the status (preliminary or final). Revised data are identified in publications but users are not informed of causes of revisions.
- No routine studies and analyses of revisions are made public; internal analysis focuses on errors and shortcomings reported by banks.

---

### 5 Accessibility

### Data accessibility and presentation
- Presentation in the monthly Bulletin of the NBU facilitates interpretation: covers money markets, balance sheets of the NBU and commercial banks, money supply, credits, with charts and tables.
- Estimates are produced only for internal analytical purposes and are not disseminated.
- No seasonal adjustment of monetary data is currently disseminated; studies on methods are under way.
- Monetary data are disseminated in hardcopy and electronic formats; monthly time series within each year (since 1992) are disseminated on the NBU website. More detailed data appear in the monthly Bulletin.
- There is no publicly announced ARC for the monthly Bulletin or NBU website dissemination, but the Economic Department follows an internal publication schedule to publish the bulletin about 30 days after the end of the reference month; users are aware of and expect this schedule.
- Data are released simultaneously to all interested users, except for cases specified elsewhere (1.2.2). Publication of the monthly Bulletin and posting on the Internet are simultaneous.
- Unpublished but nonconfidential sub-aggregates are made available free of charge upon request with authorization of senior Economic Department staff; if substantial demand exists, dissemination requires permission of the Vice Governor.

### Metadata accessibility
- The Economic Department publishes in the monthly Bulletin a short description of terms used in monetary statistics tables (sectors, instruments, aggregates, net positions).
- This description does not provide full metadata on framework, concepts, scope, accounting conventions, nature of data sources, and compilation practices. No documentation on differences from internationally accepted standards and good practices is available.
- As Ukraine intends to subscribe to the SDDS, the Economic Department prepared base page information for SDDS-recommended financial-sector data categories and sent it to the IMF for review.

### Assistance to users
- Contact information (phone numbers and e-mail addresses) of NBU officials responsible for compilation of statistical data is publicized in the monthly Bulletin.
- No brochures have been produced to educate users.
- Publications, documents, and other services are generally free; private sector subscribers of the monthly Bulletin of the NBU and Visnyk of the NBU are charged for subscription. A catalogue of publications is available on the NBU’s website and subscription transactions are executed through post offices or editorial staff.

---

*Source: _cr03256 - 3.3 Assessment and validation of source data*

### 0.2 Resources

### 0.2 Resources

### 0.2.1 Staff, financial, and computing resources are commensurate with statistical programs

- The Balance of Payments Department (BOPD) staff size: 24 employees.
- Main outputs and frequencies:
  - Monthly current account estimates (for monitoring purposes only).
  - Quarterly balance of payments estimates.
  - Experimental international investment position (IIP) statistics intended for annual publication; quarterly IIP produced for monitoring purposes.
- Additional responsibilities:
  - Selected BOPD staff participate in preparation of balance of payments forecasts and macroeconomic analysis.
  - Compilers maintain three databases for balance of payments and IIP compilation.
- Training and staff management:
  - Regular training: visits to balance of payments compiling central banks organized for groups of three to five compilers.
  - Opportunity to attend internal training offered by the NBU.
  - Staff retention: high.
- Computing and IT:
  - Each staff member is provided with a LAN-connected computer, replaced, on average, every two to three years.
  - Banks’ reports are electronically transmitted and processed using a combination of in-house and off-the-shelf software with automatic edits for source data from banks and enterprises.
- Resource adequacy and risks:
  - Human resources currently allocated for balance of payments and IIP compilation are adequate.
  - Imminent decision to disseminate annual IIP—and the possibility that compilation and dissemination of quarterly external debt data would be assumed by either the BOPD or the Foreign Exchange Regulation Department of the NBU—would stretch these resources.
- Cross-agency resource constraints:
  - Other agencies compiling selected source data face resource problems.
  - The SSCU cannot undertake a travelers’ survey due to lack of resources; such a survey would improve estimates of travel and shuttle trade.

### 0.2.2 Measures to ensure efficient use of resources are implemented

- Budgeting and staff assessment:
  - NBU budgeting: annual, by the board of directors based on departmental needs; budgets approved by the NBU Council; allocations made by the Financial Department.
  - NBU Board Resolution of June 21, 2000, No 255 “On the Attestation Procedure of NBU’s Civil Servants”: NBU staff will be subject to performance assessment and review of job description once in 3-5 years.
- Cost-saving and burden-reduction initiatives:
  - Electronic reporting by banks reduces manual data entry requirements and response burden for banks.
  - Automatic edits implemented in early phases of data processing.
  - Improved NBU website quality, including detailed balance of payments information, reduces need for paper publications and facilitates user access.

### Quality-related implications tied to resources (excerpted context)

- Planned methodological change: ITRS threshold for separately identifying individual transactions will be reduced to US$50,000 from US$250,000—expected to improve classification accuracy into balance of payments components.
- Limitations linked to resource constraints:
  - Lack of a travelers’ survey limits accuracy of travel, shuttle trade, compensation of employees, and workers’ remittances estimates.
  - No entry for reinvested earnings of FDI enterprises because imputed flows measured using SSCU’s quarterly FDI survey were not materially significant (US$1.5 million in 2001).

*Source: _cr03256 - 0.2 Resources*

### 3.1 Source data

### 3.1 Source data

### Source data collection and coverage
- The ITRS established by the NBU is a closed-bank reporting system supplemented by direct reporting of enterprises that have a bank account abroad, as well as by recourse to survey and administrative data.
- Source data provide a broadly adequate coverage of international transactions of Ukraine.
- Banks report monthly transactions according to the standard BPM5 classification, by currency of denomination, country of transaction, and country of correspondent bank.
- Banks must classify all transactions, except those below the Hryvnia equivalent of US$20,000 whose purposes do not have to be clarified; very few transactions are reported as unclassified.
- Banks are required to provide a copy of the payment order for transactions over a US$250,000 threshold; payment order information is used to monitor classification accuracy, but:
  - The US$250,000 threshold is relatively high and negatively impacts classification accuracy.
  - Payment order information does not allow identification of the resident party to the transaction.
- The Economic Department has suggested:
  - Reducing the threshold for transactions requiring additional information about the resident party.
  - Identifying the resident party using the Unified State Register of Enterprises and Organizations.
- Direct reporting of enterprises: approximately 100 enterprises must complete form 2–PB, which provides monthly information on all transactions carried out through specified accounts of Ukrainian enterprises.
- Coding of direct-reporting transactions is performed by staff of the Economic Department.

### Current account — goods, services, income, and current transfers
- Goods:
  - Trade in goods is largely derived from SCCU data compiled monthly from customs declarations and supplemented by SSCU survey information on goods procured in ports and trade on energy.
  - BOPD adjusts customs-based data to include repairs on goods, shuttle trade, and private imports of vehicles not included in customs data.
  - Source data for private imports of vehicles: customs information on number of cars imported by individuals and taxes collected.
  - Shuttle trade imports are derived using an annual commodity survey of open-air retail outlets conducted by the SSCU and estimated ratios of shuttle trade imports by commodity type.
  - BOPD staff relies on bilateral exchanges with compilers in the Russian Federation and Belarus to adjust exports for shuttle trade.
  - Methodology could be improved with a travelers’ survey; such a survey has been designed but not conducted due to lack of SSCU resources.
  - Detailed supply and use tables compiled by the National Accounts Department of SSCU would benefit shuttle trade estimates; implementation is included in SSCU’s long-term plan but detailed tables do not exist.
- Services:
  - Trade in services is largely derived from the ITRS.
  - ITRS data are supplemented by a freight adjustment (c.i.f. to f.o.b.) using results of a 2000 SSCU survey on freight and insurance payments on imports of goods.
  - SSCU provides survey information on receipts for pipeline services provided to nonresidents.
  - Travel component: BOPD uses SSCU quarterly survey on international services (tourist firms and hotels), ITRS information, and immigration statistics (number of persons crossing the border, origins, and assumptions on purposes) to estimate travel receipts and expenses; these estimates are compared for monitoring and adjustment.
  - Provision of technical assistance (sub-component of other business services) is estimated using administrative data from the Ministry of Economy.
  - SSCU’s quarterly international services survey covers 6,000 enterprises and provides export-of-services estimates close to those from the ITRS; SSCU survey imports figures are lower than ITRS, indicating superior ITRS coverage.
  - At present, there is no sharing with SSCU of the list of enterprises that import services; after amending the ITRS to identify resident parties, NBU will be able to provide such lists to SSCU to update its business register.
- Income:
  - Income transactions are largely derived from the ITRS.
  - Adjustments to income on debt (interest) liabilities use MoF information on official external debt and guaranteed debt to record on an accrual basis.
  - Reinvested earnings of direct investment enterprises are not included because they are not materially significant.
  - Dividends of direct investment enterprises are recorded net of withholding tax, not gross as recommended in BPM5; BOPD staff agreed to examine these data to report gross and record withholding tax as a current transfer.
- Current transfers:
  - Worker’s remittances derived from ITRS.
  - Technical assistance based on Ministry of Economy administrative data; recourse to administrative data is important because Ukraine receives a significant amount of technical assistance.

### Capital and financial account
- Capital account — capital transfers:
  - BOPD works closely with MoF staff responsible for official external debt and receives information on debt forgiveness occurring in debt-restructuring agreements; such debt forgiveness is recorded under capital transfers.
- Financial account — direct investment:
  - Direct investment statistics derived from ITRS and quarterly FDI survey conducted by SSCU.
  - The FDI survey covers all direct investment enterprises (registration mandatory) and most direct investors, though investor coverage is less complete.
  - ITRS supplemented with survey information for investment in kind and selected privatization receipts not captured by survey information.
  - Survey forms should be improved to identify loans granted by direct investors to direct investment enterprises (FDI investor loans).
- Portfolio and other investment:
  - Transactions on debt securities adjusted to an accrual basis using MoF information: detailed schedule of payments of interest and principal and actual disbursements for all components of official external debt.
  - This information is used to record external debt transactions—whether portfolio investment (bonds and notes) or other investment (loans)—on an accrual basis and arrears to other investment; other liabilities; general government; short-term component of the financial account.
  - ITRS information on equity securities liabilities for other sectors is regularly monitored against State Committee on Securities and Stock Exchange data or other alternative sources to confirm classification and amounts.
  - ITRS component other investment; other assets/liabilities; other sectors is supplemented by SSCU survey data on inter-enterprise receivables and payables. This monthly survey covers approximately 55,000 enterprises and provides separate information on stocks of receivables and payables with the rest of the world, broken down by CIS and other countries.
  - Needed improvements to the inter-enterprise survey are discussed in section 3.1.2.
- Reserve assets:
  - Data provided by the Foreign Exchange Regulation Department of the NBU.
  - The department recently implemented the IMF’s Data Template on International Reserves and Foreign Currency Liquidity (Reserve Template), a mandatory requirement for SDDS subscription.
  - Data compiled according to the Guidelines for a Data Template but not yet reviewed by the IMF; these data have not yet been disseminated to the public.
- Financial press is regularly monitored to review publicly available information on significant current and financial account transactions for more specific analysis.

### Identified source-data issues and suggested improvements (section 3.1.2)
- Major issues requiring action:
  - (i) Lower the threshold for banks to provide payment order information and ensure payment orders identify the resident party to the transaction.
  - (ii) Regularly conduct a travelers’ survey.
  - (iii) Revise inter-enterprise receivables/payables survey forms to include information on debt instruments, currency of denomination, debt maturity, and write-offs for bad debts.
  - (iv) Separately identify FDI investor loans in FDI survey forms.
- Despite issues, the range of data sources allows compilers to make adjustments to broadly conform to BPM5 guidelines.

### Timeliness of source data (section 3.1.3)
- NBU receives timely ITRS data from all reporting banks and enterprises.
- Timely reporting for SSCU surveys is ensured by direct follow-ups by staff of the over 600 local statistical offices and 27 regional offices, guaranteeing almost 100 percent response rate for enterprise surveys; however, some business registers are not current, leading to some under coverage.

### Statistical techniques and compilation practices (section 3.2)
- ITRS is a fully computerized system monitoring opening and closing account balances of reporting banks monthly on an individual-bank basis; discrepancies are automatically investigated and followed up.
- Banks’ balances are cross-checked with report forms used to compile IIP information; the closed ITRS enables verifications.
- The system automatically identifies large transactions, compared with payment order information for transactions above reporting threshold.
- NBU is reducing the payment-order threshold from US$250,000 to US$50,000 and will require banks to provide the enterprise unique identification number with payment orders, allowing BOPD compilers to verify reported information with the resident party.
- New bank report forms and instructions were submitted for discussion with banks’ representatives at a seminar “On the Role of the Banking System in Establishing an Information Base for the Compilation of Ukraine’s Balance of Payments”.
- Shuttle trade methodology uses all available sources; BOPD staff note estimates could be improved by a sample survey of travelers at selected border points.

### Assessment, validation, and quality control (sections 3.3–3.4)
- Source data (ITRS, surveys, administrative records) are routinely assessed:
  - Monthly comparisons of reported account balances in the closed ITRS with banks’ balances reported for money and banking statistics.
  - Equality between debit and credit entries for neutral transactions is monitored.
  - ITRS information is compared with other data sources during quarterly BOP compilation; applies to services by standard components and FDI income/equity transactions using SSCU survey results.
  - Large transactions are monitored to verify classification accuracy.
- Intermediate data validation:
  - High-value financial account transactions monitored using financial press.
  - High-value portfolio investment transactions cross-checked with State Committee on Securities and Stock Exchange or other sources to monitor reported share prices.
  - Travel data from ITRS and survey information are compared with model-based estimates using immigration counts and assumptions on purpose and average daily expenses; travel estimates would benefit from travelers’ survey implementation.
  - ITRS data on FDI income and financial flows regularly monitored using FDI survey information.
- Discrepancies and monitoring:
  - Net errors and omissions are constantly monitored; movements trigger data reviews.
  - Bilateral comparisons with individual CIS countries are done for merchandise trade.
  - NBU examined BIS locational international banking statistics and found amounts reported for Ukraine to be negligible.

### Revision studies and transparency (section 3.5)
- Materially significant revisions are highlighted using footnotes describing source and magnitude.
- BOPD staff regularly conduct analysis of revisions, but there are no published studies reviewing preliminary and final data to identify systematic sources of errors or omissions.
- Data users would benefit from detailed studies of the impact of the annual revision process and recourse to new data sources to enhance transparency.

### Serviceability — relevance, timeliness, periodicity (section 4)
- Relevance:
  - No formal consultation process with government users, private sector, or academia; no users’ survey.
  - Inter-agency working group (established by Cabinet of Ministers resolution on national account of August 9, 2001) allows BOPD staff to monitor data relevance for SNA compilers and obtain views of major data users.
  - BOPD staff participation in international meetings/seminars (e.g., IMF-sponsored 2001 CPIS and seminars at central banks of Austria, Germany, France, Japan) promotes relevance and methodology improvements.
- Timeliness and periodicity:
  - Ukraine’s balance of payments meets SDDS timeliness requirements; compiled and disseminated quarterly.
  - Data first released on NBU website approximately 75 to 80 days after the end of the reference quarter; Balance of Payments publication available later with more detailed tables and notes.
  - Merchandise trade data disseminated monthly by SSCU no later than 45 days after the end of the reporting month; disseminated on a cumulative basis, not as discrete monthly series.
  - Official reserve assets disseminated within one week after the end of the reference month.
  - The monthly Reserves Template and the annual IIP are compiled for internal NBU purposes but are not disseminated.

*Source: _cr03256 - 3.1 Source data*

### 4.3 Consistency

### _cr03256 - 4.3 Consistency

### 4.3 Consistency — Key findings
- Concepts, definitions, and classifications for quarterly and annual balance of payments statistics are the same and broadly consistent with the recommendations included in BPM5.
- Quarterly series are benchmarked to the level of annual series when annual data sources replace quarterly estimates; annual revisions take place the fourth quarter of every year.
- The current year’s estimates and quarterly data for the previous year are subject to revisions at the time of the annual revision process.
- The net errors and omissions component was negative and large during 1997–99 but significantly decreased in 2000 and 2001, suggesting improved coverage of international transactions.
- Time series for the balance of payments are consistent since 1994 — the first year the NBU started compiling and disseminating data.
- Component details for services transactions were increased in 2000 and carried back to 1999, consistent with the two-year revision policy of the NBU.
- Major revisions or exceptional transactions are explained using footnotes to statistical tables, data confidentiality permitting; analytical text provides analysis of recent fluctuations.

### 4.3.3 Cross-dataset consistency and identified issue
- Balance of payments data are consistent with other macroeconomic datasets of Ukraine with one exception:
  - Balance of payments are a direct input to the external sector of the national accounts, but final annual revisions to the national accounts take place approximately 12 months after the end of the reference year, while those of balance of payments are made 15 months after the end of the reference year.
  - The authorities should address the difference in revision schedules for national accounts and balance of payments, which leads to inconsistencies between the two datasets.
- Merchandise trade data are compiled on a cumulative basis during the year; monthly trade data are revised once a year in time for the revised annual balance of payments estimates.
- Money and banking statistics and official external debt statistics are consistent with balance of payments.

### 4.4 Revision policy and practice — Key findings and recommendations
- Revision cycle:
  - Quarterly data are not revised until the publication of the fourth quarter data; at that time all quarters of the current and the previous year are subject to revision.
  - The revision cycle is long-established but not clearly made known to the public; draft metadata describing the revision policy and schedule have not been posted on the NBU’s website or included in NBU publications.
- Documentation and implementation:
  - Any change in methodologies is documented in the sources and methods section of the quarterly Balance of Payments publication.
  - Implementation of new data collection procedures follows an established schedule; e.g., consultation with banks on revising the threshold for detailed payment orders in form 1-PB, with implementation following a board resolution specifying an effective date.
- Preliminary data:
  - The NBU website notes that quarterly data are provisional while time series data are “refined.”
  - Recommendation: The Economics Department should better inform the public that quarterly data are preliminary and revised once a year (at fourth-quarter publication) and once more 12 months later.
- Revision studies:
  - Statistical tables and website footnotes highlight and explain major revisions.
  - The BOPD does not prepare detailed analysis of revisions by standard components (including reasons and sources); producing such analyses would enhance transparency.

### 5 Accessibility — Data dissemination and user access
- Presentation:
  - The NBU’s balance of payments publication follows the standard component presentation and includes detailed statistical tables, aggregated and detailed time-series presentations for the rest of the world and two main economic regions, and supplementary tables on exports/imports by major categories and by country.
  - Analytical presentation of the balance of payments is not published; separate tables showing the impact of debt rescheduling agreements are not provided.
- Coordination with SSCU:
  - The NBU publishes adjustments made to SSCU trade data; the SSCU notes its merchandise trade estimates exclude shuttle trade.
- Timeliness:
  - Data are first released on the NBU website approximately 75 to 80 days after the end of the reference quarter; this is followed by a publication with analysis and supplementary tables.
- Advance release calendar:
  - An advance release calendar (ARC) is not provided. (Footnote: Following Ukraine’s subscription to the SDDS, an ARC is regularly updated on the NBU and SSCU websites.)
- Simultaneous release:
  - Data are released simultaneously to all users via the NBU website.
- Nonpublished sub-aggregates:
  - Few nonpublished sub-aggregates are available. NBU collects data by country and could provide more geographical breakdowns, but the BOPD cannot respond to such requests because work on a cost-recovery basis is not a practice adopted by the NBU and would be too costly to fulfill.
  - Additional information on external debt, including exceptional financing transactions, could be provided upon request to the Economic Department.
  - Supplementary details on direct investment by region of Ukraine, country of investor, industry, and type of transaction are available upon request to the SSCU, but these data are not fully consistent with the balance of payments because the latter are supplemented by banks’ reporting and privatization receipts when relevant.
  - Additional details for merchandise trade on a customs basis are available upon request to the SSCU; SSCU processes requests on a cost-recovery basis when applicable.

### 5.2 Metadata accessibility and assistance to users
- Documentation:
  - The note on sources and methods in the quarterly balance of payments publication documents concepts, scope, classifications, basis of recording, data sources, and statistical techniques.
  - The note does not specify that inter-enterprises’ receivables and payables are not broken down into trade credits, FDI loans, and others, nor that investment income is recorded net of withholding tax and not gross, as recommended in the BPM5.
- User support:
  - The quarterly balance of payments publication lists senior compilers and their areas of specialization with telephone and facsimile numbers; staff provide service and support in Ukrainian, Russian, and English.
  - The NBU website provides lists of statistical publications in Ukrainian and English; the quarterly balance of payments publication is available in Ukrainian and in a bilingual English-Russian format. Data on the NBU website are available in Ukrainian and English.

### Identified thresholds, timing, and numerical details
- Time series start: 1994.
- Net errors and omissions large and negative: 1997–99; decreased in 2000 and 2001.
- Annual revision timing: fourth quarter of every year.
- National accounts final annual revisions: approximately 12 months after the end of the reference year.
- Balance of payments final annual revisions: 15 months after the end of the reference year.
- Data release lag: approximately 75 to 80 days after the end of the reference quarter.
- Bank reporting threshold currently prevents monitoring below US$250,000; NBU is expected to implement a US$50,000 threshold.
- Quarterly data revised at fourth-quarter publication and once more 12 months later.

### Table 6 — Data Quality Assessment Framework: notable assessments (summary)
- Areas marked as needing attention or "LNO/NO" include:
  - 1.2 Transparency: "There is no publicly available information on access to statistics prior to release. There is no advance notice of major change in methodology."
  - 2.2 Scope: "There is no travelers’ survey for shuttle trade and travel estimates. The coverage for compensation of employees is insufficient."
  - 2.3 Classification/sectorization: "Component details for current transfers are not provided. No breakdown is provided for inter-enterprises’ receivables and payables between trade credits, FDI loans, and other instruments."
  - 3.1 Source data: "The reporting threshold for banks to provide payment order information to the NBU is too high and does not allow monitoring the accuracy of the classification of transactions below US$250,000 (the NBU is expected to shortly implement a US$50,000 threshold). The survey form used to collect data on inter-enterprises’ receivables and payables should be reviewed to include information on debt instruments, currency of denomination, debt maturity, and write-off of bad debts."
  - 3.5 Revision studies: "The results of revision studies are not disseminated. These studies should provide users with information on systematic sources of errors or omissions."
  - 4.1 Relevance: "There is no users’ survey to assess the relevance of existing statistics."
  - 4.3 Consistency: "Balance of payments and national accounts are inconsistent because of different revision policies. The net errors and omissions component was high during 1997–99."
  - 4.4 Revision policy and practice: "Information on the revision policy is not available to the public. The BOPD does not publish documents on sources and magnitude of the revisions and explain the impact of recourse to new data sources."
  - 5.1 Data accessibility: "The ARC is not made available to the public. Few additional details provided on balance of payments."

*Source: _cr03256 - 4.3 Consistency (Balance of Payments Statistics, compiling agency: National Bank of Ukraine).*

### 3. Accuracy and

### 3. Accuracy and reliability

### Overview
- Source data and statistical techniques are sound and statistical outputs sufficiently portray reality.
- Quality dimensions addressed: source data, statistical techniques, assessment and validation, and revision studies.

### Source data (Element 3.1)
- 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.
- Summary indicator: Source data available provide an adequate basis to compile statistics.

### Statistical techniques (Element 3.2)
- 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.
- Summary indicator: Statistical techniques employed conform to sound statistical procedures.

### Assessment and validation (Elements 3.3–3.4)
- 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.
- 3.4.1 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.
- Summary indicators:
  - 3.3 Assessment and validation of source data – Source data are regularly assessed and validated.
  - 3.4 Assessment and validation of intermediate data and statistical outputs – Intermediate results and statistical outputs are regularly assessed and validated.

### Revision studies (Element 3.5)
- 3.5.1 Studies and analyses of revisions are carried out routinely and used to inform statistical processes.
- Summary indicator: Revisions, as a gauge of reliability, are tracked and mined for the information they may provide.

---

### 4. Serviceability

### Overview
- Statistics are relevant, timely, consistent, and follow a predictable revisions policy.
- Quality elements: relevance, timeliness and periodicity, consistency, revision policy and practice.

### Relevance (Element 4.1)
- 4.1.1 The relevance and practical utility of existing statistics in meeting users’ needs are monitored.
- Summary indicator: Statistics cover relevant information on the subject field.

### Timeliness and periodicity (Element 4.2)
- 4.2.1 Timeliness follows dissemination standards.
- 4.2.2 Periodicity follows dissemination standards.
- Summary indicator: Timeliness and periodicity follow internationally accepted dissemination standards.

### Consistency (Element 4.3)
- 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.
- Summary indicator: Statistics are consistent within the dataset, over time, and with major datasets.

### Revision policy and practice (Element 4.4)
- 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.
- Summary indicator: Data revisions follow a regular and publicized procedure.

---

### 5. Accessibility

### Overview
- Data and metadata are easily available and assistance to users is adequate.
- Quality elements: data accessibility, metadata accessibility, assistance to users.

### Data accessibility (Element 5.1)
- 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.
- Summary indicator: Statistics are presented in a clear and understandable manner, forms of dissemination are adequate, and statistics are made available on an impartial basis.

### Metadata accessibility (Element 5.2)
- 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.
- Summary indicator: Up-to-date and pertinent metadata are made available.

### Assistance to users (Element 5.3)
- 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.
- Summary indicator: Prompt and knowledgeable support service is available.

---

### Users’ Survey (Ukraine) — summary findings

Survey context
- Thirty responses were received, including 11 from universities/research institutes, 1 commercial bank, 5 private sector users, and 13 government agencies.
- Respondents: 13 Government, 11 Universities/Research Institutes, 1 Bank, 5 Private Sector, TOTAL 30.

General perceptions
- Respondents’ views on Ukraine’s statistics were generally positive.
- Highest satisfaction: overall quality of the statistics, underlying methodology, periodicity of the data, and access.
- Largest dissatisfaction: level of detail and coverage of the statistics, availability of information about revisions, and methodological descriptions.

Major user concerns and requests
- Nonavailability of data in time series format (fixed reference period) for GDP, CPI, PPI, and industrial production index.
- Publication of data on a cumulative basis (e.g., monthly merchandise trade) is not appropriate.
- Requests for more detailed breakdowns of the SNA components and improvements to GDP deflation methodology.
- After annual revisions, users want revised underlying quarterly series to be provided.
- Demand for compilation of input-output tables at greater commodity/industry detail and implementation of a set of financial accounts.
- Recognition that rapid implementation of requested improvements requires additional resources for SNA compilation within the State Statistics Committee.
- Appreciation of availability of monthly GDP and regional accounts statistics.
- No consensus on trade-off between timeliness and accuracy/reliability of national accounts; some prefer longer dissemination lags for accuracy, others prefer current compromise but more frequent revisions.
- Support for adoption of international standards and classifications, with concern over absence of retrospective adjustments after methodological changes (affecting GDP, industrial production index, merchandise trade, and GFS).
- Need for better explanations of sources and impact of large revisions to balance of payments statistics.
- Electronic data availability upon special request is satisfactory, but SSCU website lacks up-to-date information and does not allow retrieval of data in electronic format; NBU’s website cited as the best statistical website of Ukraine.
- Absence of up-to-date methodological notes for many macroeconomic series (including GDP and GFS); difficulty reconciling GFS disseminated by MoF with government sector data in SNA—users requested reconciliation items.
- Requests for reconciliation items between balance of payments and merchandise trade statistics.
- Monetary statistics viewed as best set of macroeconomic statistics, but users request more detailed methodological notes and detailed data on interest rates from NBU.
- Concern about absence of coordination of revision policies for national accounts and balance of payments; recent establishment of an inter-agency working group on national accounts seen as beneficial.
- Absence of a national body to consider all statistical problems as a whole and advise on priorities for improvements identified as a problem.

Survey quantitative results (selected items)
- Uses of official macroeconomic statistics (number of respondents indicating regular use):
  - 1.1 National accounts: 21
  - 1.2 Production indices: 23
  - 1.3 Prices: 22
  - 1.4 Labor market: 22
  - 1.5 Balance of payments: 19
  - 1.6 Merchandise trade: 22
  - 1.7 International reserves and foreign liquidity: 13
  - 1.8 External debt: 17
  - 1.9 International investment position: 12
  - 1.10 Government/public sector operations and debt: 16
  - 1.11 Monetary and financial statistics: 22
  - 1.12 Other: 14

- Sources for national official statistics (number of respondents):
  - 2.1 Official press releases and publications on macroeconomic statistics: 27
  - 2.2 Private sector summaries and analyses: 3
  - 2.3 Official policy papers: 25
  - 2.4 Publications from international organizations about the country: 17
  - 2.5 Other: 7

- Reference to official descriptions of sources and methods:
  - YES: 23
  - NO: 7

- Purposes for use (number of respondents indicating use):
  - 4.1 Analysis of current developments for short-term decision making: 21
  - 4.2 Analysis of trend for longer-term policy formulation: 21
  - 4.3 Econometric model building and forecasting: 18
  - 4.4 Economic research: 20
  - 4.5 Comparison with other countries: 16
  - 4.6 General economic background: 22
  - 4.7 Other: 6

- Coverage and detail:
  - 5.1 In general, are you satisfied with the coverage of official statistics?
    - YES: 14
    - NO: 15
    - NO OPINION: 1
  - 5.2 Satisfied with level of detail?
    - YES: 11
    - NO: 18

- Periodicity and timeliness:
  - 6.1 Satisfied with frequency of compilation?
    - YES: 23
    - NO: 5
    - NO OPINION: 1
  - 6.2 Consider dissemination timeliness appropriate?
    - YES: 17
    - NO: 11
    - NO OPINION: 1

- Dissemination practices:
  - 7.1 Know if a public release calendar exists?
    - YES: 13
    - NO: 17
  - 7.2 If calendar exists, are releases on announced dates?
    - YES: 9
    - NO: 6
    - NO OPINION: 14
  - 7.3 Is there enough information about revisions?
    - YES: 6
    - NO: 15
    - NO OPINION: 6

- Accessibility:
  - 8.1 Can you easily access official statistics identified in question 1?
    - YES: 19
    - NO: 11
  - 8.2 Can you easily access explanatory notes and methodological descriptions?
    - YES: 13
    - NO: 14
    - NO OPINION: 1
  - 8.3 Is methodology information sufficiently clear and detailed?
    - YES: 14
    - NO: 11
    - NO OPINION: 4
  - 8.4 How do you get access to official statistics? (selected response counts)
    - Official releases: 18
    - Publications (hard copy): 27
    - Data specifically requested: 16
    - Official website: 13
    - Electronic (E-mail): 7

---

*Source: Data Quality Assessment Framework—Generic Framework (July 2001 Vintage) and Users’ Survey results as provided in the source PDF.*

### 9. Overall assessment

### 9. Overall assessment

### Survey sample
- Number of Respondents: Government 13, Universities/Research Institutes 11, Banks 1, Private Sector 5, TOTAL 30

### 9.1 Underlying methodology sound and appropriate?
- YES: 3 7 1 3 14
- NO: 2  1  3
- NO OPINION: 7 4  1 12

### 9.2 Official statistics unbiased and accurate?
- YES: 4 9 1 3 17
- NO: 7  1 8
- NO OPINION:  2 1 3

### 9.3 Overall quality of the official statistics?
- 1 - poor: 0
- 2: 2
- 3: 9 2  2 13
- 4: 6 7 1 1 15
- 5 - excellent:  1 1 2

### Areas for improvement — Coverage
- Prices (2)
- Measurement of active population
- GDP deflators (2)
- Finance statistics
- External trade (2)
- Payrolls
- Paid services by type
- Households' incomes and expenditures
- Regional statistics
- Use of natural resources
- Environmental protection
- Small businesses
- Investment
- Structure of debt in arrears

### Areas for improvement — Level of detail
- More detailed industrial breakdowns
- Sufficient level of detail by industry, especially breakdown of industrial production by type of activity
- Regional accounts (3)
- Business finance
- Labor remuneration
- Classification of traditional and new industries
- Budget appropriations to healthcare institutions by region
- SNA
- Statistical reporting by selected agricultural businesses
- Value added structure by quarters
- Investment structure

### Areas for improvement — Periodicity
- SNA
- Price indices
- State Treasury reports
- Regional statistics should be produced monthly
- Annual periodicity of data release is adequate
- Quarterly supply of data is enough

### Areas for improvement — Timeliness
- Compilation lag for revised annual GDP should be less than 12 months after reference year
- Compilation period should be reduced to 30 days (2)
- GDP for December and the fourth quarter
- Sales of goods including by barter
- Financial results of enterprises

### Release calendar and revisions
- Provision of an advance release calendar:
  - No information about such calendar
  - No public calendar of revisions to data series
- Provision of information about revisions:
  - Revisions of annual GDP are not always accompanied by revisions of the quarterly indicators
  - Explanations on large revisions are not always available (2)
  - Official statistics of Ukraine are rarely revised due to the very high quality of the data

### Access to official statistics
- No access to detailed breakdowns and revisions
- Statistics are mainly disseminated in the forms of press releases
- Insufficient data on the state budget execution
- Relatively easy access to the data although some indicators are not released publicly
- There is no easy access to external debt data and SNA
- Required data are supplied in electronic format on special request (3)

### Access to metadata and methodological notes
- The SSCU website must include methodological notes (now available only by request)
- Not enough details about compilation methods for GDP assessment
- Metadata are often insufficient or stale
- Cannot find the methodological papers

### Methodological issues
- Require more discrete data versus cumulative data
- SNA does not use a fixed base year for constant price GDP components
- Require improvements to the quality of GDP deflators
- Labor market data are biased
- NBU and SSCU use different methodologies for trade in goods statistics compilation
- Computation of fixed assets is inadequate
- Measurement of the shadow economy is an issue (2)
- Lack of import/export prices

### Assessment of Ukraine’s statistics (user comments)
- As good as those of the other CIS countries (3) but lagging behind Eastern Europe (1)
- Equal to that of Eastern Europe
- Difficult to make the judgment
- Russian statistics are better
- Quality of the statistics is high although the ongoing revisions bring some uncertainty about accuracy of the data
- Ukrainian statistics are of high quality and do not undergo frequent revisions
- Average quality statistics but better than Russian or Polish
- Adequate quality statistics
- Good quality
- Ukraine's statistics are one of the best within the CIS
- Quality of the statistics is quite satisfactory for Ukraine
- Quality of the official statistics is insufficient

### Suggestions (user recommendations)
- Improve statistical methodologies (3)
- Provide better access to the SSCU compilation methods for individual indicators (2)
- Provide better access to statistics (6), especially in electronic format (2) and for budgetary spending units (1)
- Provide more information on the data revisions policy
- Show statistical discrepancies for GDP assessed by different approaches
- Improve compilation methods for GDP deflators (2)
- Provide more detailed breakdowns of SNA (2)
- Provide consistency between the NBU and SSCU estimates for merchandise trade (3)
- Explain the links between government sector in the SNA and state budget reporting (4)
- Improve estimates of the shadow economy
- Make available IIP data
- Improve consistency between production by industrial branches and value added data (2)
- Improve compilation methods for the index of industrial production
- Improve coverage of enterprises of all types of ownership
- Improve all areas of real sector statistics: industrial breakdowns, incomes and expenditures, value added, profit, employment, labor efficiency
- Compute gross value added for the regions, cities, territories, urban agglomerations, economic regions, border zones
- Compile regional statistics for all data categories (2)
- Disseminate data on the state debt by maturity
- Improve quality of the employment data by region

*Source: Ukraine — Users’ Survey (responses and comments summarized).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2003/_cr03256.pdf_