## wp18175

## Source details

**Canonical URL:** [wp18175](https://www.imf.org/-/media/files/publications/wp/2018/wp18175.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2018/wp18175.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2018/wp18175.pdf.json)

---

### Introduction
- Administrative data can significantly enhance the timeliness, detail, and accuracy of national statistics, supporting quicker and better-targeted policy responses.
- Administrative data are increasingly accessible due to rapidly advancing technology and can be adapted for statistical use at lower cost than new data collection.
- Administrative sources beyond tax data—health, education, welfare, demographic, social security, migration—can be merged with tax data to enrich economic statistics and analysis.

### Using tax data to enhance national accounts
- Roles of tax records:
  - Support annual and quarterly national accounts and a monthly index of economic activity (MIEA).
- VAT data:
  - Useful for tracking new industries and business models and for estimating activity in industries with small GDP shares.
  - VAT turnover or sales may be good proxies for output (assuming unchanged inventories).
  - Enterprise-level monthly VAT data provide sales and purchases; purchases can proxy intermediate consumption.
- Corporate income tax records:
  - When financial statements are available, can be used to compile detailed production accounts, income accounts, financial accounts, and balance sheets of non-financial corporations.
- Financial sector tax records:
  - Less useful for national accounts because financial activity is often VAT-exempt and tax records do not straightforwardly capture net interest income.

### Data sharing and confidentiality
- Risks:
  - Careless or undocumented sharing can undermine tax authority confidence and tax compliance; many jurisdictions use informal exchanges despite risks.
- Mechanisms to protect confidentiality:
  - Partially anonymized data (identification numbers instead of names; inclusion of general characteristics).
  - Restricted access to detailed records for a limited group of NSO staff; auditing and sanctions for breaches.
  - Secure transmission protocols to address external risks (hacking).
- Legal frameworks and MoUs:
  - Legislation is preferable to MoUs; some countries (UK, South Africa) operate “data laboratories”.
  - MoUs can specify that data are shared for statistical purposes only, schedules, formats, harmonization, and security measures.

### Cooperation between tax authorities and NSOs
- Mutual benefits:
  - NSOs gain data for statistics; tax authorities can benefit from NSO assistance with data quality and ISIC industry coding and from joint data security arrangements.
  - Joint comparison of tax registries and business registers can detect errors and unregistered taxpayers.
- MoUs:
  - Promote reliability and continuity; can include procedures for harmonization, schedules, formats, and senior-level engagement.
  - Two sample MoUs referenced: Uganda (2008) and The Gambia (signed January 6, 2016).
- Data validation:
  - Establish robust validation and quality checks; concentrate detailed validation on larger companies when full validation is impractical.

### Coverage of the tax base — key benchmark and empirical findings
- General rule of thumb: registration and on time filing rates should be at least 60 percent.
- RA-FIT findings (revenue administration surveys covering fiscal years 2011–15):
  - Across three income groups, average taxpayer registration rates exceed 60 percent for corporate, individual income, and VAT taxes.
  - On-time VAT tax return filing rates exceed 60 percent across groups.
  - On-time corporate tax filing rates exceed 60 percent only for LICs and UMIs; general improvement from 2011 to 2015 noted.
  - Note: on-time plus delayed filing rates could be substantially higher.
- Specific averaged rates (as presented in the source):
  - Income tax on corporations and other entities
    - Low income: 75.0, 76.3, 74.2, ..., ...
    - Lower middle income: 64.5, 63.5, 71.5, ..., ...
    - Upper middle income: 68.0, 68.0, 70.2, ..., ...
  - Individual income tax
    - Low income: 76.7, 74.6, 75.3, ..., ...
    - Lower middle income: 76.3, 75.9, 69.8, ..., ...
    - Upper middle income: 66.2, 64.3, 64.1, ..., ...
  - VAT tax
    - Low income: 85.3, 87.7, 85.1, ..., ...
    - Lower middle income: 71.7, 68.3, 72.8, ..., ...
    - Upper middle income: 66.2, 65.0, 71.5, ..., ...
  - On-time filing rates — VAT Returns
    - Low income: 67.7, 64.2, 71.0, 73.4, 76.7
    - Lower middle income: 62.7, 63.9, 66.4, 76.4, 77.2
    - Upper middle income: 71.7, 72.1, 70.2, 75.3, 73.1
  - On-time filing rates — Corporate income tax returns
    - Low income: 64.1, 67.3, 62.3, 74.4, 76.0
    - Lower middle income: 46.5, 47.6, 51.1, 58.5, 58.1
    - Upper middle income: 48.3, 52.7, 58.3, 62.0, 61.7
- Empirical scope:
  - Examination of 90 countries during 2011–2015 finds VAT and corporate income tax registration and filing rates adequate for national accounts compilation purposes for almost all countries across different income groups.

### Rough edges (limitations and adjustments)
- A. Variations in compliance
  - Filing obligations, frequency, information requirements, and registration thresholds can change over time.
  - Tax compliance may vary; changes in taxpayer registries or revenues must be separated into changes reflecting economic output versus reporting behavior.
  - Tax evasion can be negatively correlated with GDP growth and distort taxable income–GDP relationships.
  - Tax waivers (particularly for new investors) can result in production that is not immediately captured by tax records.
  - Tax data can represent a biased sample; calibration against independent data sources is recommended.
- B. Time of recording
  - Tax data are often recorded on a cash basis while national accounts use accrual accounting; timing mismatches complicate value added calculations.
  - Industries with better alignment: mining and quarrying, much of manufacturing, hotels and restaurants, transportation, communications.
  - Industries with poor alignment: construction, retail/wholesale where sales and purchases are recorded at different times.
  - Other timing issues: non-reporting in some periods, late reporting, frequency mismatches, seasonality, and different reporting periodicities.
- C. Classification
  - NSOs use ISIC (International Standard Industrial Classification); tax authorities may use informal or less structured classifications.
  - Large enterprises with multiple activities may misclassify themselves; harmonization of coding is important.
- D. The informal sector
  - Generally not captured by tax data; NSO estimates of informal activity can inform tax authorities about potential compliance gaps and represent an opportunity for collaboration.

### Taxpayer registries and business registers
- Taxpayer registries:
  - Maintained by tax authorities to monitor compliance; should have legislative basis, unique identifiers, and update procedures.
- Business registers:
  - Maintained by NSOs to guide data collection and sample selection; record ISIC classifications, entry/exit, enterprise sizes.
- Consolidation and coordination:
  - Possibility to consolidate or better coordinate taxpayer registries and NSO business registers to avoid duplication.
  - Software tools can assist in matching and deduplicating registries.

### Use of tax data in national accounts — country experiences (selected examples)
- Countries using VAT and income tax records in compiling MIEA, quarterly and annual GDP, and business registers include: Canada, Chile, Denmark, Finland, Guatemala, Pacific Island countries, Rwanda, Uganda, United Kingdom, United States.
- Country highlights:
  - Canada: GST data used to replace sub-annual surveys for simple establishments; GST reported monthly for large companies, quarterly for average-sized, annually for small; Unified Enterprise Survey (UES) replaced sub-annual surveys for many small enterprises; over 50 percent of simple businesses stopped receiving UES questionnaires.
  - Chile: VAT returns used for IMACEC components, quarterly production accounts for some industries, and income tax returns for annual accounts; tax microdata also used in sectoral accounts.
  - Denmark: longstanding register-based censuses and broad use of administrative data; 2011 official census cost about US$150,000.
  - Finland: VAT and PAYE data used extensively; enterprises with turnover over EUR 8,500 covered; 93 percent of enterprises accounting for an estimated 99.8 percent of turnover report VAT monthly.
  - Guatemala: Tax records used for components of the MIEA since 2007; informal data-sharing with the Tax Administration Office enabled quarterly national accounts.
  - Pacific Island countries: Monthly VAT reporting by registered businesses can yield useful indicators; MoUs and Statistics Act protections used in about half the countries.
  - Rwanda: Quarterly VAT data on 19,000 VAT reporters and 31,000 annual income tax submissions shared with NISR in a standard format each month.
  - Uganda: MoU between UBoS and URA adopted in 2008; UBoS maintains records on approximately 60,000 firms updated monthly from URA; designated Contact Officers and a Liaison Committee manage exchanges.
  - United Kingdom: Program to use HMRC VAT turnover data to supplement Monthly Business Survey; VAT turnover data from 630,000 businesses integrated into GDP estimates published December 22, 2017; plan to transform short-term turnover statistics by end of 2020.
  - United States: February 2014 OMB memo called for expanded use of administrative data and stewardship practices; 2016 Evidence-Based Policymaking Commission Act established a Commission whose final report was issued in September 2017.

### Appendix and tools for assessing coverage
- ISORA/RA-FIT, RA-GAP and TADAT:
  - International Survey on Revenue Administration: self-reported indicators including time filing rates and payment rates; fiscal years 2014 and 2015 currently available.
  - RA-FIT (Revenue Administration Fiscal Information Tool): Web-based tool developed by FAD; on time filing data for 2011–13 collected through RA-FIT; data for 2011 to 2015 aggregated by income groups and region available on the RA-FIT Data Dissemination Platform.
  - TADAT (Tax Administration Diagnostic Tool): focuses on nine key performance outcome areas (a) integrity of the registered taxpayer base, b) risk management, c) supporting voluntary compliance, d) filing of tax returns, e) payment of obligations, f) ensuring accuracy of reporting, g) tax dispute resolution, h) operational efficiency and effectiveness, i) accountability and transparency).
  - RA-GAP (Revenue Administration Gap Analysis Program): estimates potential tax revenues, evaluates actual revenues, analyses factors causing gaps; when statistical data are adequate, estimates the overall tax gap broken down into compliance gap and policy gap, with further breakdowns.

### Appendix II — Sample MoU provisions (highlights)
- Uganda Bureau of Statistics (UBOS) and Uganda Revenue Authority (URA) MOU (2008) key clauses:
  - URA shall provide UBOS timely tax statistics summarized from traders’ returns on a monthly basis.
  - UBOS will use data only for compiling aggregate statistics and drawing samples for economic surveys; will work with URA to improve data quality and processing systems.
  - Security and confidentiality: data released in anonymous format; UBOS shall not allow data to be used for non-statistical purposes.
  - Contact Officers and a liaison committee coordinate operations; regular meetings and secure transmission timetables.
  - Termination: either party may terminate by serving a three months' written notice.
- The Gambia Revenue Authority (GRA) and The Gambia Bureau of Statistics (GBoS) MOU (signed January 6, 2016) key clauses:
  - Agreement scope: coordination, sharing, updating and periodic reporting of VAT data and Business Register data; part of a framework agreement.
  - GRA will provide GBoS, on time, with VAT data: quarterly VAT data based on ISIC Rev.4 for quarterly GDP estimates and publication; updates on Business Register; Import and Export data for external trade statistics and development of XMPI.
  - GBoS will provide GRA with industry coding assistance (ISIC Rev. 4), GDP estimates, external trade statistics, Business Register updates (subject to Statistics Act 2005 confidentiality), CPI (available every 15th of the month following the reference month), PPI (available every quarter), social indicators and population projections.
  - Planning and delivery: timely exchange of electronic data, including individual records, subject to respect of statistical confidentiality; delivery follows respective production and publication timeframes.
  - Quality standards: data delivered must meet defined quality standards.
  - Use of resources: capacity and budgeting responsibilities defined in annual budgets; new workload subject to reallocation or new funding.
  - Leadership and management: defined Working Group leaders and Management Committee membership; quarterly reporting to Management Committee.
  - Costs: Both GRA and GBoS will not claim mutual costs; may share costs if needed.
  - Copyright and reproduction: copyright remains with the relevant institution; parties may reproduce data as most recently released and may aggregate or seasonally adjust without reference; consult on major modifications.
  - Confidentiality: data subject to Secrecy laws of the Income and Value Added Tax Act 2012 and the Statistics Act 2005; individual records treated as confidential.
  - Duration: effective from January 6, 2016; operation reviewed annually; signed in two copies.

### Conclusion — policy recommendations and practical actions
- Verify that registration and filing rates for taxation are adequate (benchmark: at least 60 percent on-time filing/registration).
- Address confidentiality and legal restrictions by:
  - Using partially anonymized data where appropriate.
  - Restricting access to designated NSO staff with audit and sanction mechanisms.
  - Establishing secure transmission and storage protocols.
- Formalize data sharing through concise MoUs or, preferably, statutory frameworks that:
  - Specify mutual benefits, data schedules, formats, harmonization, and confidentiality safeguards.
  - Clarify that data are used for statistical purposes only.
- Reorganize and transform administrative data for statistical use; invest in NSO capacity to convert and analyze administrative data.
- Harmonize classification systems (ISIC) and coordinate taxpayer registries with business registers to improve coverage and avoid duplication.
- Use calibration and independent data sources to address sample bias and changes in compliance and to construct quarterly estimates from annual data when needed.
- Focus validation efforts on larger enterprises when resource constraints make full validation impractical.
- Pursue mutual-benefit collaborations where NSOs can assist tax authorities (data quality, coding) and tax authorities can support NSOs (data provision, registries).

*Source: IMF staff paper — 1. Adequacy of Taxpayer Data Coverage for National Accounts Purposes (excerpts supplied).*

### References ________________________________________________________________22

### References ________________________________________________________________22

### Figures
- 1. Collaboration Between Tax Authorities and National Statistics Offices ______________10

### Tables
- Table

*Source: wp18175 - References (PDF), canonical URL: https://www.imf.org/-/media/files/publications/wp/2018/wp18175.pdf*

### 1. Adequacy of Taxpayer Data Coverage for National Accounts Purposes ______________11

### 1. Adequacy of Taxpayer Data Coverage for National Accounts Purposes ______________11

### Introduction
- Administrative data can significantly enhance the timeliness, detail, and accuracy of national statistics, supporting quicker and better-targeted policy responses.
- Administrative data are increasingly accessible due to rapidly advancing technology and can be adapted for statistical use at lower cost than new data collection.
- Administrative sources beyond tax data—health, education, welfare, demographic, social security, migration—can be merged with tax data to enrich economic statistics and analysis.

### Using tax data to enhance national accounts
- Tax records can support annual and quarterly national accounts and a monthly index of economic activity (MIEA).
- VAT data:
  - Useful for tracking new industries and business models and for estimating activity in industries with small GDP shares.
  - VAT turnover or sales may be good proxies for output (assuming unchanged inventories).
  - Enterprise-level monthly VAT data provide sales and purchases; purchases can proxy intermediate consumption.
- Corporate income tax records:
  - When financial statements are available, can be used to compile detailed production accounts, income accounts, financial accounts, and balance sheets of non-financial corporations.
- Financial sector tax records:
  - Less useful for national accounts because financial activity is often VAT-exempt and tax records do not straightforwardly capture net interest income (see FISIM note).

### Data sharing and confidentiality
- Careless or undocumented sharing can undermine tax authority confidence and tax compliance; many jurisdictions use informal exchanges despite risks.
- Mechanisms to protect confidentiality:
  - Partially anonymized data (identification numbers instead of names; inclusion of general characteristics).
  - Restricted access to detailed records for a limited group of NSO staff; auditing and sanctions for breaches.
  - Secure transmission protocols to address external risks (hacking).
- Legal frameworks and MoUs:
  - Legislation is preferable to MoUs; some countries (UK, South Africa) operate “data laboratories”.
  - MoUs can specify that data are shared for statistical purposes only, schedules, formats, harmonization, and security measures.

### Cooperation between tax authorities and NSOs
- Mutual benefits:
  - NSOs gain data for statistics; tax authorities can benefit from NSO assistance with data quality and ISIC industry coding and from joint data security arrangements.
  - Joint comparison of tax registries and business registers can detect errors and unregistered taxpayers.
- MoUs:
  - Promote reliability and continuity; can include procedures for harmonization, schedules, formats, and senior-level engagement.
  - Two sample MoUs referenced: Uganda and The Gambia (Appendix II).
- Data validation:
  - Establish robust validation and quality checks; concentrate detailed validation on larger companies when full validation is impractical.

### Coverage of the tax base — key benchmark and empirical findings
- General rule of thumb: registration and on time filing rates should be at least 60 percent.
- RA-FIT (Revenue Administration Fiscal Information Tool) findings (revenue administration surveys covering fiscal years 2011–15):
  - Across three income groups, average taxpayer registration rates exceed 60 percent for corporate, individual income, and VAT taxes.
  - On-time VAT tax return filing rates exceed 60 percent across groups.
  - On-time corporate tax filing rates exceed 60 percent only for LICs and UMIs; general improvement from 2011 to 2015 noted.
  - Note: on-time plus delayed filing rates could be substantially higher.
- Specific averaged rates (as presented in the source):
  - Income tax on corporations and other entities
    - Low income: 75.0, 76.3, 74.2, ..., ...
    - Lower middle income: 64.5, 63.5, 71.5, ..., ...
    - Upper middle income: 68.0, 68.0, 70.2, ..., ...
  - Individual income tax
    - Low income: 76.7, 74.6, 75.3, ..., ...
    - Lower middle income: 76.3, 75.9, 69.8, ..., ...
    - Upper middle income: 66.2, 64.3, 64.1, ..., ...
  - VAT tax
    - Low income: 85.3, 87.7, 85.1, ..., ...
    - Lower middle income: 71.7, 68.3, 72.8, ..., ...
    - Upper middle income: 66.2, 65.0, 71.5, ..., ...
  - On-time filing rates — VAT Returns
    - Low income: 67.7, 64.2, 71.0, 73.4, 76.7
    - Lower middle income: 62.7, 63.9, 66.4, 76.4, 77.2
    - Upper middle income: 71.7, 72.1, 70.2, 75.3, 73.1
  - On-time filing rates — Corporate income tax returns
    - Low income: 64.1, 67.3, 62.3, 74.4, 76.0
    - Lower middle income: 46.5, 47.6, 51.1, 58.5, 58.1
    - Upper middle income: 48.3, 52.7, 58.3, 62.0, 61.7
- Empirical scope: examination of 90 countries during 2011–2015 finds VAT and corporate income tax registration and filing rates adequate for national accounts compilation purposes for almost all countries across different income groups.

### Rough edges (limitations and adjustments)
- A. Variations in compliance
  - Filing obligations, frequency, information requirements, and registration thresholds can change over time.
  - Tax compliance may vary; changes in taxpayer registries or revenues must be separated into changes reflecting economic output versus reporting behavior.
  - Tax evasion can be negatively correlated with GDP growth and distort taxable income–GDP relationships.
  - Tax waivers (particularly for new investors) can result in production that is not immediately captured by tax records.
  - Tax data can represent a biased sample; calibration against independent data sources is recommended.
- B. Time of recording
  - Tax data are often recorded on a cash basis while national accounts use accrual accounting; timing mismatches complicate value added calculations.
  - Industries with better alignment: mining and quarrying, much of manufacturing, hotels and restaurants, transportation, communications.
  - Industries with poor alignment: construction, retail/wholesale where sales and purchases are recorded at different times.
  - Other timing issues: non-reporting in some periods, late reporting, frequency mismatches, seasonality, and different reporting periodicities.
- C. Classification
  - NSOs use ISIC (International Standard Industrial Classification); tax authorities may use informal or less structured classifications.
  - Large enterprises with multiple activities may misclassify themselves; harmonization of coding is important.
- D. The informal sector
  - Generally not captured by tax data; NSO estimates of informal activity can inform tax authorities about potential compliance gaps and represent an opportunity for collaboration.

### Taxpayer registries and business registers
- Taxpayer registries:
  - Maintained by tax authorities to monitor compliance; should have legislative basis, unique identifiers, and update procedures.
- Business registers:
  - Maintained by NSOs to guide data collection and sample selection; record ISIC classifications, entry/exit, enterprise sizes.
- Consolidation and coordination:
  - Possibility to consolidate or better coordinate taxpayer registries and NSO business registers to avoid duplication.
  - Software tools can assist in matching and deduplicating registries.

### Use of tax data in national accounts — country experiences (selected examples)
- Countries using VAT and income tax records in compiling MIEA, quarterly and annual GDP, and business registers: Canada, Chile, Denmark, Finland, Guatemala, Pacific Island countries, Rwanda, Uganda, United Kingdom, United States.
- Canada:
  - GST data used to replace sub-annual surveys for simple establishments; GST reported monthly for large companies, quarterly for average-sized, annually for small.
  - Unified Enterprise Survey (UES) replaced sub-annual surveys for many small enterprises; over 50 percent of simple businesses stopped receiving UES questionnaires.
- Chile:
  - VAT returns used for IMACEC components, quarterly production accounts for some industries, and income tax returns for annual accounts; tax microdata also used in sectoral accounts.
- Denmark:
  - Longstanding register-based censuses and broad use of administrative data for multiple statistics; 2011 official census cost about US$150,000.
- Finland:
  - VAT and PAYE data used extensively; enterprises with turnover over EUR 8,500 covered; 93 percent of enterprises accounting for an estimated 99.8 percent of turnover report VAT monthly.
- Guatemala:
  - Tax records used for components of the MIEA since 2007; informal data-sharing with the Tax Administration Office enabled quarterly national accounts.
- Pacific Island countries:
  - Monthly VAT reporting by registered businesses can yield useful indicators for national accounts; MoUs and Statistics Act protections used in about half the countries.
- Rwanda:
  - Quarterly VAT data on 19,000 VAT reporters and 31,000 annual income tax submissions shared with NISR in a standard format each month.
- Uganda:
  - MoU between UBoS and URA adopted in 2008; UBoS maintains records on approximately 60,000 firms updated monthly from URA; designated Contact Officers and a Liaison Committee manage exchanges.
- United Kingdom:
  - Program to use HMRC VAT turnover data to supplement Monthly Business Survey; VAT turnover data from 630,000 businesses integrated into GDP estimates published December 22, 2017; plan to transform short-term turnover statistics by end of 2020.
- United States:
  - February 2014 OMB memo called for expanded use of administrative data and stewardship practices; 2016 Evidence-Based Policymaking Commission Act established a Commission whose final report was issued in September 2017.

### Conclusion — policy recommendations and practical actions
- Verify that registration and filing rates for taxation are adequate (benchmark: at least 60 percent on-time filing/registration).
- Address confidentiality and legal restrictions by:
  - Using partially anonymized data where appropriate.
  - Restricting access to designated NSO staff with audit and sanction mechanisms.
  - Establishing secure transmission and storage protocols.
- Formalize data sharing through concise MoUs or, preferably, statutory frameworks that:
  - Specify mutual benefits, data schedules, formats, harmonization, and confidentiality safeguards.
  - Clarify that data are used for statistical purposes only.
- Reorganize and transform administrative data for statistical use; invest in NSO capacity to convert and analyze administrative data.
- Harmonize classification systems (ISIC) and coordinate taxpayer registries with business registers to improve coverage and avoid duplication.
- Use calibration and independent data sources to address sample bias and changes in compliance and to construct quarterly estimates from annual data when needed.
- Focus validation efforts on larger enterprises when resource constraints make full validation impractical.
- Pursue mutual-benefit collaborations where NSOs can assist tax authorities (data quality, coding) and tax authorities can support NSOs (data provision, registries).

*Source: IMF staff paper — 1. Adequacy of Taxpayer Data Coverage for National Accounts Purposes (excerpts supplied).*

### REFERENCES

### wp18175 - REFERENCES

### References (selected)
- Allcoat, John, “Methodology: HMRC VAT update April 2016”, Office for National Statistics, United Kingdom, April 4, 2016.
- Allingham, Michael Agnar and Sandmo, “Income Tax Evasion: A Theoretical Analysis”, Journal of Public Economics, 1972, 323-338.
- Bean Charles, Independent Review of UK Economic Statistics, March 2016.
- Bloem Adriaan, Robert Dippelsman, and Nils O. Mӕhle, “Quarterly National Accounts Manual, Concepts, Data Sources, and Compilation”, 2001.
- Bissett D., Peter, “Use of Tax Data in the Production of Provincial Economic Statistics”, Statistics Canada, 5/21/2014.
- Brodeur, Marie, Use of Tax Data in the Unified Enterprise Survey (UES), Statistics Canada Workshop on Use of Administrative Data in Economics Statistics, Moscow, October 30th–November 1st, 2006.
- Commission of the European Communities, International Monetary Fund, Organisation for Economic Co-operation and Development, United Nations, and World Bank, “System of National Accounts 2008”, New York, December 2009.
- Duncan Cleary, William Crandall, and Andrew Masters, “Understanding Revenue Administration, Results from the Second Survey of the Revenue Administration—Fiscal Information Tool”, Fiscal Affairs Department, International Monetary Fund, 2017.
- Fernández, Ivette, “Use of Tax Micro-data for the Compilation of the Chilean National Accounts”, IFC Bulletin No. 37 165, Central Bank of Chile.
- Freeman, Russel, The Use of Administrative Data in the Pacific Island Countries, LTX News, IMF, November 2015 (unpublished).
- Grote, Martin, “How to Establish a Tax Policy Unit”, How-To-Note No. 17/02, IMF, October 2017.
- Lemgruber, Andrea and Andrew Masters and Duncan Clary, “Understanding Revenue Administration, An Initial Data Analysis Using the Revenue Administration Fiscal Information Tool”, Fiscal Affairs Department, International Monetary Fund, 2015.
- OECD, “Improving Co-operation Between Tax and Anti-Money Laundering Authorities”, September 2015.
- Paavilainen, Paula, “Efficient Use of Administrative Data in the Production of Economic Statistics in Finland”, Statistics Finland Business Trends, FIN-00022, Statistics Finland, Finland, 2012.
- Pier, Louis and Marie Brodeur, “Statistical Use of Goods and Services Tax Data in Statistics Canada’s Monthly Economic Surveys”, Statistics Canada.
- Statistics Canada, “Non-technical Background Paper on the Unified Enterprise Survey Program”, June 1999.
- Statistics Denmark, “Administrative Data in the Production of Official Statistics—An Initiative on How to Enhance Data Coverage in Relation to Sustainable Development Goals and Targets (SDG) and the National Statistical Systems in General”, March 3, 2016.
- Statistics Denmark, “Issues and Challenges Faced in Countries when Seeking to Increase Use of Administrative Data”, 2017.
- Stephens, Mark and John Allcoat, “Exploitation of HMRC VAT data, Office for National Statistics”, United Kingdom, October 7, 2015.
- Stephens, Mark, VAT Turnover Implementation into National Accounts: December 2017 Update, Office for National Statistics, December 22, 2017.
- Uganda Bureau of Statistics, “Memorandum of Understanding (MoU), between UBoS and Uganda Revenue Authority”, 2008.
- United Nations, “International Standard Industrial Classification of All Economic Activities (ISIC)”, Rev. 4, 2008.
- U.S. Office of Management and Budget, “Memorandum for the Heads of Executive Departments and Agencies”, February 2014.

### Appendix I: Tools Providing an Indication of the Coverage of Tax Records: ISORA/RA-FIT, RA-GAP and TADAT
- International Survey on Revenue Administration
  - Purpose: gathering information on revenue administration practices and performance to support comparative studies and benchmarking.
  - Coverage: tax authorities self-report on indicators including time filing rates and payment rates.
  - Data availability: fiscal years 2014 and 2015 currently available.
- RA-FIT (Revenue Administration Fiscal Information Tool)
  - Web-based tool developed by FAD.
  - On time filing data for 2011–13 were collected through RA-FIT.
  - Data for 2011 to 2015, aggregated by income groups and region, are available publicly on the associated RA-FIT Data Dissemination Platform (http://data.rafit.org).
  - Income groups for 2011–13 divided according to 2012 Gross National Income (GNI) per capita, calculated using the World Bank Atlas method.
  - Income groups for 2014–15 updated by the World Bank are based on 2013 data.
- TADAT (Tax Administration Diagnostic Tool)
  - Multidimensional diagnostic tool aimed at identifying tax administration reform priorities.
  - Focuses on nine key performance outcome areas:
    - a) integrity of the registered taxpayer base
    - b) risk management
    - c) supporting voluntary compliance
    - d) filing of tax returns
    - e) payment of obligations
    - f) ensuring accuracy of reporting
    - g) tax dispute resolution
    - h) operational efficiency and effectiveness
    - i) accountability and transparency
- RA-GAP (Revenue Administration Gap Analysis Program)
  - IMF technical assistance service assisting tax administrations in monitoring taxpayer compliance levels through tax gap analysis.
  - Functions:
    - estimates potential tax revenues
    - evaluates actual revenues
    - analyses factors causing gaps between potential and actual revenues
  - When statistical data are adequate, estimates the overall tax gap broken down into:
    - compliance gap and policy gap
    - compliance gap can be further broken down into a collections gap and an assessment gap
    - policy gap into a non-taxable gap and an expenditure gap
  - Purpose: improve effectiveness in raising revenue, promote perceived fairness among taxpayers, and build trust in the tax system.

### Countries included in Table 1 (as listed)
- Albania, Anguilla, Argentina, Barbados, Belize, Benin, Bermuda, Bolivia, Bosnia and Herzegovina, Brazil, Bulgaria, Burkina Faso, Burundi, Cameroon, Cape Verde, Central African Republic, Chad, Chile, Colombia, Republic of Congo, Democratic Republic of Congo, Cook Islands, Costa Rica, Cote d'Ivoire, Dominica, Dominican Republic, Ecuador, El Salvador, Ethiopia, Fiji, Gabon, The Gambia, Ghana, Grenada, Guatemala, Guinea, Guinea-Bissau, Honduras, Ireland, Jamaica, Jordan, Kenya, Kiribati, Kosovo, Lesotho, Liberia, Macedonia, Madagascar, Malawi, Maldives, Mali, Marshall Islands, Mauritania, Mauritius, Mexico, Micronesia, Montenegro, Montserrat, Netherlands, Nicaragua, Niger, Nigeria, Norway, Papua New Guinea, Paraguay, Peru, Philippines, Rwanda, Samoa, Senegal, Serbia, Seychelles, Sierra Leone, Solomon Islands, St. Kitts and Nevis, St. Lucia, St. Vincent and the Grenadines, Sudan, Swaziland, Tanzania, Togo, Tonga, Trinidad and Tobago, Turks and Caicos Islands, Uganda, Uruguay, Vanuatu, Virgin Islands, West Bank and Gaza, and Zambia.

### Appendix II: Sample Memoranda of Understanding (MOU) — Key clauses and provisions
- Uganda Bureau of Statistics (UBOS) and Uganda Revenue Authority (URA) MOU (2008)
  - Preamble: formalise procedures relating to the flow of accurate and timely domestic tax data for compilation of gross domestic product (in current and constant prices).
  - Definitions: includes terms such as "Agreement", "Chief Contact Officers", "Contact Officers", "URA", "UBOS", “VAT”, “PAYE”.
  - Paragraph 2 — General Responsibilities:
    - URA shall provide to Uganda Bureau of Statistics timely Tax statistics on a regular basis as summarized traders’ returns on a monthly basis.
    - UBOS undertakes to use the data supplied for purposes of: Only compiling aggregate statistics needed for the monitoring and analysis of economic developments and for drawing of samples of businesses necessary for conducting economic surveys
    - UBOS on the basis of data supplied shall work with URA to improve data quality and where necessary make improvements to sourcing and processing systems to the benefit of both parties.
    - UBOS undertakes to consult URA on any requirements that might be included in surveys conducted by UBOS and provide any other statistical advice whenever called upon.
  - Paragraph 3 — Security and Confidentiality:
    - UBOS shall not allow the data supplied by URA to be used for any other purposes other than for statistical inferences.
    - Any data to be released shall be in an anonymous format in order to up hold the confidentiality of the affairs of the taxpayers.
  - Paragraph 4 — Contact Officers:
    - UBOS and URA shall have contact officers whose names and official titles shall be communicated to either party.
    - Contact officers shall form a liaison committee to coordinate operations and meet within a week after data exchange in alternate locations and with alternate chairpersons.
  - Paragraph 5 — Transmission of Information:
    - URA shall supply UBOS with the data according to regular timetables, using secure modes of data exchange and in formats that will be agreed upon between the parties.
    - URA shall advise UBOS of any changes or up dates to the information handling and processing deemed important to its use.
  - Paragraph 6 — Accuracy and Quality:
    - URA shall carry out normal processes including data editing needed to ensure that Tax statistics are of acceptable quality.
    - UBOS shall monitor quality and immediately inform URA of any discrepancies.
    - UBOS and the URA shall hold regular meetings to review data and statistics compiled during processing.
    - UBOS shall to any extent possible provide the statistics generated to URA in an agreed format before their release to the general public.
  - Paragraph 7 — Amendments:
    - Any amendment to the arrangement shall be mutually agreed and in writing.
  - Paragraph 8 — Miscellaneous Provisions:
    - A party shall not re-assign obligations without written consent of the other party.
    - Information obtained from URA by UBOS under this understanding shall extend to any other information as of when statistical need arises.
  - Paragraph 9 — Settlement of Disputes:
    - URA AND UBOS shall co-operate to resolve amicably any disagreement or dispute arising out of the interpretation or execution of this memorandum.
  - Paragraph 10 — Termination and Continuity:
    - Either party may terminate by serving a three months' written notice.
    - Termination takes effect at the end of the three months from date of service.
    - The terminating party shall continue to perform outstanding obligations.
  - Paragraph 11 — Force Majeure:
    - Operations suspended if parties prevented by civil strife or unforeseeable natural and technological calamities.
  - Paragraph 12 — Commencement:
    - Agreement shall come into effect on the date of signing by both parties.
  - Signatures: signed in 6 copies; date indicated as "At ...........................................on the............................day of.......................2008."
- The Gambia Revenue Authority (GRA) and The Gambia Bureau of Statistics (GBoS) MOU (signed January 6, 2016)
  - Agreement scope: coordination, sharing, updating and periodic reporting of VAT data and Business Register data; part of a framework agreement between GRA and GBoS.
  - Considerations:
    - Recognition of the need to enhance coordination of the National Statistical System in The Gambia and use of VAT data for quarterly GDP estimates and development of the Business Register.
    - GBoS is the state-mandated agency for coordinating the National Statistical System and publishing quarterly GDP estimates and maintaining the Business Register.
    - GRA is mandated to collect taxes and has technical expertise and an extensive database of taxpayers with Tax Identification Numbers (TIN).
    - GBoS and GRA will work together to classify businesses according to “International Standard Industrial Classification (ISIC Rev 4)”.
  - Article 1 — Purpose:
    - The MOU is not a legally binding contract; intended to provide clear understanding of services, roles, responsibilities, and procedures for resolving disagreements.
  - Article 2 — Role of parties and working arrangements:
    - Joint working group responsibilities include design and implementation of the Business Register, rules for matching and reconciling business entities, birthing and deathing procedures, industry coding, and updating variables; analyse VAT data over time according to ISIC Rev 4 and assess reliability for quarterly GDP estimates.
    - GRA will provide GBoS, on time, with:
      - VAT data: quarterly VAT data based on International Standards Industrial Classification (ISIC Rev.4) for quarterly GDP estimates and publication in the quarterly statistics abstract.
      - Updates on Business register: information needed to develop and maintain the Business Register.
      - Import and Export data: external trade statistics (including exports and imports data) for compilation of external trade statistics quarterly report and development of Export and Import Price Index (XMPI).

*wp18175 - REFERENCES*

### 3. The Gambia Bureau of Statistics (GBoS) will provide GRA on time with the

### 3. The Gambia Bureau of Statistics (GBoS) will provide GRA on time with the

### Data provisions to GRA
- Industry Coding: GBoS will assist GRA in industry coding on ongoing basis using ISIC Rev. 4.
- National Accounts statistics:
  - GDP estimates will be available annually according to the release calendar of GBoS and communicated to GRA.
  - Once the quarterly GDP is developed and ready for release, it will be shared with GRA.
- External Trade statistics: quarterly external trade statistics report will be shared with GRA as well as XMPI once it is developed.
- Business Register: updated structure of the Business Register according to ISIC Rev 4 will be shared with GRA, subject to statistics confidentiality as defined by the Statistics Act 2005.
- Price statistics:
  - monthly Consumer price index (CPI): The CPI will be made available every 15th of the month following the CPI reference month.
  - producer price index (PPI): PPI will be made available every quarter.
  - All these products will be made available and submitted to the GRA.
- Social indicators: GBoS will provide GRA with the needed social indicators and population projections, gathered from surveys and censuses or other administrative sources.

### Article 3: Planning and time of delivery
- Mutual access: GBoS needs quick and reliable access to data compiled by GRA, and GRA needs similar access to data compiled by GBoS.
- Timely exchange: Parties agree for the timely exchange of electronic data, including individual records, subject to the respect of statistical confidentiality.
- Delivery timing: The delivery of the data to each institution will be done following their respective production and publication timeframes, conditioned on both institutions presenting data in time.
- Documentation: Both GRA and GBoS will discuss and draft a document that lists the agreements made.

### Article 4: Quality standards
- The data to be delivered to each institution will have to meet the defined quality standards.

### Article 5: Use of resources
- Capacity and budgeting: The needed capacity for each institution to meet this agreement will be defined in their respective annual budget.
- New workload: Where it causes significant new workload demand, the provision of the data will be subject to the department’s ability to reallocate or receive new funding for the required resources.

### Article 6: Leadership, Cooperation and Communication
- Legal representation:
  - GRA relevant leading technical unit is legally represented by the Deputy Director, Policy Planning & Research.
  - GBoS relevant leading technical unit is legally represented by the Director of National Accounts Statistics.
- Cooperation channels: bilateral contacts, participation in working meetings and committees, National Statistics Council (NSC) meetings, and the Conference of Official Statistics annual meetings by users and producers.
- Daily management: run by the Principal Statistician in charge of Business Enterprise Surveys of GBoS and the Deputy Director, Policy Planning & Research of the GRA, who will be the designated Working Group leaders for their agencies.
- Management Committee composition:
  - For GRA: the Commissioner of Domestic Taxes, Deputy Director Policy Planning & Research, Manager of GamTaxNet.
  - For GBoS: the Director of Quality, Coordination and Dissemination; the Director of National Accounts; the Director of Prices and Government Finance Statistics; the Principal Statistician in charge of Business Enterprise Surveys in National Accounts Unit.
  - Other members: the Director of Registrar of Companies, the Director of the Gambia Tourism Board, and the Director of the Chamber of Commerce.
- Reporting: At least once a quarter the Working Group leaders will report to the Management Committee.
- Dispute resolution:
  - Resolve disagreements at the Working Group level where possible.
  - If unresolved, the Commissioner of Domestic Taxes of the GRA and the Director of National Accounts of the GBoS will endeavour to resolve the issue.
  - If still unresolved, the Management Committee will endeavour to come to an agreement.
- Undefined matters: All matters not defined in this agreement will be decided upon by the Management Committee.
- Oversight: The signatories represent the institutions providing relevant statistics covered by this MOU and will oversee the implementation of the MOU.

### Article 7: Costs, invoicing, payments
- No mutual cost claims: Both GRA and GBoS will not claim any cost from each other.
- Cost sharing: The two institutions will share costs for carrying out some relevant statistical activities if need arises.

### Article 8: Copyrights, Authorship and Reproduction of data
- Copyright ownership: The copyright of the work done by both GRA and GBoS will remain with the relevant institution.
- Reproduction and modification:
  - GRA and GBoS may reproduce, in electronic or hard copy publications, data compiled by the other party.
  - In principle, both parties may reproduce the data as most recently released by the other party as a prime source.
  - The data may be aggregated, or seasonally adjusted in a different way by the party reproducing them without reference to the other party.
  - The parties shall consult each other on any other major modification to the data published.

### Article 9: Confidentiality
- Legal protections:
  - Data provided to GBoS by GRA is subject to the Secrecy laws of the Income and Value Added Tax Act 2012: information received shall be used by the GBoS officer solely for the performance of their duties, and information regarding individual records is treated as confidential and not to be disclosed to other parties.
  - Data provided to GBoS or submitted by GBoS to other agencies are subject to the protection of confidentiality of individual records or any other aggregation that can allow identification of individual records according to the Statistics Act 2005.

### Article 10: Duration of the agreement and updates
- Effective date: This MOU is effective from January 6, 2016.
- Duration and review:
  - The agreement will last as long as it may be deemed necessary.
  - Its operation will be monitored and will be reviewed annually.
  - At the request of either party, the GRA and GBoS may review and update this MOU.
- Signatures: Signed in two copies     Date: January 6, 2016
  - For the Gambia Revenue Authority (GRA): The Commissioner General
  - For the Gambia Bureau of Statistics (GBoS): The Statistician General

*Source: wp18175 - 3. The Gambia Bureau of Statistics (GBoS) will provide GRA on time with the (pdf).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2018/wp18175.pdf_