## _wp05218 ## Source details **Canonical URL:** [_wp05218](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp05218.pdf) ## Other formats - [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp05218.pdf.md) - [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp05218.pdf.json) --- ### I. Introduction - IMF’s Fiscal Affairs Department (FAD) provided substantial technical assistance in implementing and improving value-added tax (VAT) systems in developing and transitional countries over the past two decades. - VAT is a key component of the tax system in over 130 countries, raising about 25 percent of the world’s tax revenue. - This working paper follows up on refunding VAT excess credits based on a survey of tax administrations in 36 developing, transitional, and developed countries. - Survey requested information in four key areas: - (1) general information on VAT systems (registration threshold, VAT rates, zero-rating, number of VAT payers); - (2) details of VAT refund systems (statutory provisions, categories of refund recipients, number and size of refund claims, claimant procedures, organizational arrangements, specific regimes); - (3) data on tax audits and identification of refund fraud; - (4) evaluation and proposals for improvement. ### II. Overview — key features and problems - Invoice-credit VAT feature: - Businesses sometimes pay more VAT on purchases than they collect and should reclaim the difference; exporters with zero-rated exports commonly generate excess credits. - Magnitude and regional patterns: - In many countries, VAT refund levels exceed 40 percent of gross VAT collections. - Forty percent of survey respondents repay a third or more of gross VAT collections in refunds. - Countries with refund levels below 20 percent are mostly in Africa, Asia, and Latin America. - Timeliness and administrative practice: - Developed countries: refunds generally paid within four weeks of a refund claim. - Developing/transitional countries: processing often takes several months, sometimes more than a year; delays harm exporter competitiveness and working capital. - Causes of delays and safeguards: - Fraudulent claims prompt time-consuming verification checks by less advanced administrations, causing backlogs. - Delays also occur when state budgets are under pressure or tax collection targets are unmet. - Administrations with forecasting and budgeting capabilities can predict refund levels with fair precision. - Legal frameworks and remedies: - 90 percent of IMF survey respondents reported tax authorities are bound by law to make refunds within a prescribed timeframe, generally 30 days. - Around 40 percent of surveyed countries provide for interest on late refunds. - In 60 percent of surveyed countries, mandatory carry-forward periods for excess VAT credits are imposed, generally for nonexporters. - Alternative arrangements and trade-offs: - Zero-rating exports and deferral/exemption for imported capital goods reduce refund claims but add complexity and revenue risks. - Large-scale cross-checking and VAT bank account schemes lock working capital and increase compliance costs. - Denial of credits for large cash purchases unless paid through banks is used in some countries. - Kenya requires CPA certification for large refund claims. - Growing trend: fast-track refund processing for taxpayers with proven good compliance records. ### III. Country experiences with VAT refunds — summary findings - Refunds can be substantial in absolute terms and as a share of collections. - Refund levels vary widely across regions and tend to be higher in advanced and emerging economies. - Within regions, countries with similar VAT systems and conditions show similar refund levels. - Refund patterns within countries are relatively constant year-to-year. - Refund claims are dominated by exporters. - Most countries have statutory deadlines for refunds, but these are frequently not met. - More than half the countries surveyed do not provide interest on late refunds. - All countries report VAT refund abuse but most cannot reliably estimate associated revenue losses. - Some administrations focus audit resources mainly on refunds, potentially neglecting other VAT fraud/evasion risks. ### IV. Size of refund claims — regional and country statistics (selected) - Table 1. Value of VAT Refunds by Country/Region (In percent of gross VAT collections) — Average1 - Canada 50.3 - EU 38.1 - Eastern Europe 36.8 - New Zealand 35.5 - Former Soviet Union countries 29.6 - Latin America 17.4 - Middle East 16.2 - Asia (not including Singapore) 7.0 - Africa (not including South Africa) 6.0 - 1 Average refund level over a four-year period (1998 to 2001). - Table 2. Value of VAT Refunds in Advanced, Transitional, and Emerging Economies (In percent of gross VAT collections) — Average1 - Advanced Economies: Canada 50.3; France 21.2; Ireland 24.9; Netherlands 50.0; New Zealand 35.5; Sweden 48.6; United Kingdom 40.9 - Transitional Economies: Bulgaria 21.5; Hungary 48.2; Latvia 49.1; Romania 24.7; Russia 44.6; Slovak Rep. 53.9; Ukraine 24.1 - Emerging Economies: Chile 28.8; Colombia 4.1; Indonesia 12.4; Mexico 32.1; Morocco 5.1; South Africa 39.5 - Others: Algeria 24.3; Bolivia 10.4; Cambodia 2.8; Cameroon 8.8; El Salvador 9.6; Kenya 7.2; Mozambique 2.7; Peru 19.8 - 1 Average refund level over a four-year period (1998 to 2001). ### V. Determinants of refund levels and a simple identity - Factors influencing refund level (percent of gross VAT collections): - (1) Nature of the economy (investment-induced excess credits, export sector value-added, share of taxable vs zero-rated sales). - (2) VAT system design (zero-rating extent, multiple rates). - (3) Taxpayer compliance behavior and VAT fraud prevalence. - (4) Tax administration system and culture (corruption level, capacity to detect/prevent fraud, service commitment). - Simple identity for refunds under a fully functioning, single-rate VAT: - (α I + β(1 – λ)Z)/ẽ - Definitions: - I and Z = shares of investment and zero-rated items (including exports) in GDP; - α = proportion of investment generating excess credits; - β = proportion of zero-rated sales generating excess credits; - λ = ratio of value added to sales in zero-rated sector; - ẽ = gross efficiency ratio (gross VAT collections as a percent of GDP per percentage point of tax). - Example: - I = 10 percent; Z = 40 percent of GDP; α = 5 percent; λ = 40 percent; β = 1; ẽ = 0.9 → refunds = 27 percent of gross collections. ### VI. The Empirical Model (Box 1) — equation and main findings - Estimated equation: - Refunds = 0.16 Exports + .75 Growth + .19 Literacy + .90 Range – 25.3 D1 + 3.8 D2 – 17.5 D3 - (2.06)* (0.69) (3.41)* (2.6)* (-2.51)* (0.83) (-3.70)* - t-ratios in parentheses; * denotes significance at 5 percent. - Adjusted R-squared = 0.8826. - Variable definitions: - Refunds = average refunds to gross VAT collections over 1998-2001. - Exports = share of exports in GDP (percent). - Growth = average GDP growth rate (percent). - Literacy = literacy rate (percent). - Range = difference between highest and lowest (nonzero) VAT rates (percent). - D1 = 1 if supplies to exporters are zero-rated. - D2 = 1 when refunds are paid from gross collections (rather than appropriation). - D3 = 1 for “other” economies in Table 2 (outliers with weak refund performance). - Main empirical findings: - Refunds rise with openness (Exports), significant at 5 percent. - Refunds fall with less mature administrations (D3), significant at 5 percent. - Growth coefficient positive but not significant. - Literacy positive and significant. - Range of VAT rates positive and significant. - D1 (zero-rating exporters) lowers refund ratio (negative, significant). - D2 (paying refunds out of general revenue) has positive coefficient (not significant in this equation) — sign robust across specifications; suggests appropriations may retard refund payments. ### VII. Refund recipients and composition (Box 1 — section B) - Exporters dominate refund claims in number and value. - Typical claimant categories: - Exporters (often the majority of value). - Registered domestic zero-rated suppliers (e.g., hospitals, universities). - Traders with temporary excess credits (seasonal slumps). - Start-ups or firms with large capital purchases. - Traders under dual-rate structures. - Traders subject to withholding arrangements. - Nonregistered claimants (diplomats, tourists). - Exporters’ share of refund claims (selected 2001 examples; shares in percent of total claims): - Cameroon: number 60; value 66. - Kenya: number 70; value 48. - Morocco: number 80; value 80. - Cambodia: number 76; value 41. - Slovak Republic: number 56; value 63. - Kazakhstan: number 100; value 100. - Bolivia: number 100; value 100. - Chile: number 64; value 89. - Colombia: number 53; value 63. - El Salvador: number 100; value 100. - Peru: number 100; value 100. ### VIII. Processing times, statutory rules, and administrative practices - Statutory deadlines: - 90 percent of respondents report legal refund timeframes. - Range: 24 hours (Peru, where security provided by claimant) to 90 days (France); most common = 30 days (40 percent of countries). - France’s code stipulates 90 days, but authorities apply administrative standard of 30 days. - Interest on late refunds: - Around 40 percent of surveyed countries provide interest on late refunds. - De minimis rules exist (example: Singapore and the United Kingdom). - Interest often aligned to commercial bank rates and adjusted quarterly or half-yearly. - Carry-forward rules: - 60 percent of surveyed countries require carry-forward of excess credits, generally 3 to 6 months, range 30 days to more than a year. - EU Sixth Directive allows Member States to choose refund or carry-forward. - Ireland, the United Kingdom, Sweden, and the Netherlands reported no mandatory carry-forward periods. - Offsetting refunds against other tax liabilities: - VAT laws in 80 percent of surveyed countries allow offsetting against other tax debts. - Practices vary on timing and scope; some former Soviet states include nontax debts. - Canada and Hungary allow offsetting against customs duties. - Administrative thresholds and extraordinary measures: - Some countries do not process claims below a threshold (Italy, France, Peru). - At least two African countries legislated to deny payment of all outstanding refund claims as of a set date to wipe backlogs (practice criticized for long-term consequences). ### IX. VAT refund abuse and revenue loss estimates (Box 1 — section D) - VAT refund abuse causes substantial revenue loss through error, deliberate understatement, and systemic attacks. - U.K. estimates (2002–03 context): - VAT losses estimated around 15.8 percent of net VAT receipts in 2002–03 (fraud and nonfraud losses). - 2002–03 U.K. VAT receipts were £108.8 billion, of which £45.2 billion was refunded. - Table 4. Estimate of VAT Revenue Losses in the United Kingdom in 2001–02 (in £ billion): - Noncompliance by traders: 2.5 to 4.0 - VAT “missing trader” fraud: 1.77 to 2.75 - Traders failing to register: 0.4 to 0.5 - VAT avoidance schemes: estimate not provided in excerpt - Example enforcement action: - In 2002–03 U.K. authorities disallowed "£63 million" of VAT refunds to exporters suspected of carousel fraud. - U.K. target: reduce VAT gap from "15.8 percent" in "2002–03" to "12 percent" by "2005–06". ### X. Risk management, information exchange, and audit capacity - Risk assessment uptake: - Two-thirds of surveyed countries undertake risk-assessments in refund processing; scope varies widely. - A quarter report a statutory requirement to verify every refund claim prior to payment. - A third reported no VAT audit program; pre-refund audits dominate in a quarter of countries. - Reasons for weak audit programs: - Insufficient skilled auditors and remuneration. - Concerns about collusion between taxpayers and auditors. - Inadequate preparation at VAT implementation. - Lack of political support and weak legal/judicial environment. - Information exchange: - Unified administrations exchange VAT and income tax data automatically; separate agencies often have limited exchange (examples: Egypt, Tanzania). - VAT-customs information exchange exists in all surveyed countries (case-by-case to routine monthly data to on-line access). - Five countries had VAT and customs administered by same entity: Canada, Mexico, Peru, South Africa, United Kingdom. - EU: VIES for intra-Community supplies; U.K. had bilateral agreements with eight member states (three more in pipeline). ### XI. Budgeting and organizational arrangements for refunds - Less than "40 percent" of surveyed countries make specific expenditure appropriation for VAT refunds. - Two budgeting methods: - Pay from gross VAT revenue ("60 percent" of respondents). - Pay from budget expenditure appropriations. - Organizational arrangements: - "47 percent" of survey countries have dedicated units for processing VAT refunds. - Others process refunds within broader returns-processing or audit operations. - Some use central automated processing centers (Netherlands, New Zealand, United Kingdom). - Countries reluctant to adopt self-assessment more often establish dedicated refund units and verify claims prior to payment. ### XII. Administering refunds in a self-assessment environment and compliance strategy - Principles for self-assessment environment: - Taxpayers calculate liabilities and refunds; administration assists and enforces. - Need clear laws/procedures, resources, compliance programs mixing education, assistance, enforcement, verification, and enforcement tools. - Compliance strategy design: - Apply risk-management principles; balance revenue protection with not overburdening compliant taxpayers. - Well-developed program identifies major risks, responses, and evaluation criteria. - Example U.K. strategy elements: - Make it easy for legitimate traders to pay VAT. - Make VAT fraud difficult for dishonest traders. - Challenge abusive avoidance schemes through courts. - Detect unregistered trading. - Identify traders not paying correct VAT. ### XIII. Detecting, investigating, and preventing fraud — practices and audit design - Fraud types summarized (Box 2): - Inflated refund claims; underreported sales; fictitious traders; domestic sales disguised as exports; missing trader intra-EU fraud and carousel variants; nonregistration; credit claimed on exempt/private purchases; credit for invoices from unregistered suppliers; illegal imports with VAT not remitted; barter arrangements. - Audit program design features: - Broad coverage by size and sector; spread audit resources across program elements. - Limit pre-refund audits to high-risk cases; use selective post-refund audits for lower risk. - Short, issue-oriented VAT audits; focus on accounting systems for large taxpayers. - Coordinate VAT audit with income tax audits. - Apply consistent penalties and pursue serious fraud prosecutions. ### XIV. Specific approaches to VAT refund problems (Section V; Schemes A–G) - A. Zero-Rating Supplies to Exporters - Eliminates exporter refund claims but breaks VAT credit chain, shifting control problems and adding complexity and revenue risk. - Examples: France, Ireland, Netherlands (historical), Azerbaijan. - B. Large-Scale Cross-Checking of Invoices - Computerized invoice matching (Korea, China, Bulgaria, Indonesia, Azerbaijan, Albania) can identify false invoices but imposes high administrative and taxpayer costs; often generates substantial unproductive work. - C. Certification of Refund Claims by CPAs (Kenya) - CPA certification required for large claims; Kenyan officials reported a 40 percent drop in exporter refund claims after introduction. - Effective only with high-integrity accounting profession and acceptance of added compliance costs. - D. Preferential Treatment of Good Compliers (“Gold” Status; Pakistan example) - Categorize claimants (Gold/Silver/Others); Gold claims approved without pre-refund audit within 3-5 days; Silver subject to brief review within 15 days with assigned upper refund limit. - Post-refund audits and withdrawal of status for persistent inaccuracies. - Reduces administrative costs and accelerates refunds for compliant exporters. - E. Payment for Large Purchases Through Banking System - Require bank payments above thresholds (France, Hungary, Turkey) or deny credits for cash purchases (Azerbaijan). - Can reduce unreported transactions but may be disruptive and less effective for unregistered traders. - F. VAT Bank Accounts (Bulgaria) - VAT bank accounts freeze VAT funds and require deposits for VAT-heavy transactions. - Compliance costs high: frozen working capital, doubled payment/administration tasks, disproportionate impact on SMEs; benefits not proven. - G. Deferral of VAT on Imported Capital Goods - Registered VAT importers defer accounting for VAT liability until next return, offsetting input credit against liability; avoids immediate cash payment and refund wait. - Aims to reduce refund claims and address investor cash-flow concerns. ### XV. Comparative evaluation of schemes (Table 6 summary) - Performance criteria: Reduces delay | Reduces number of claims | Enhances VAT base protection | Reduces taxpayer compliance costs | Saves admin resources - A. Zero-rated supplies to exporters: Yes | Yes | No | No | No - B. Large-scale invoice matching: No | No | Yes | No | No - C. CPA certification: Yes | No | Yes | No | Yes - D. Preferential treatment of good compliers: Yes | No | Yes | Yes | Yes - E. Purchases paid through banking system: Yes | No | Yes | No | No - F. VAT bank accounts: No | No | Not proven | No | No - G. Deferment of VAT on capital goods: Yes | Yes | No | No | No ### XVI. Suggested model of best practice — key features and recommendations - Regulate number of VAT taxpayers via realistic registration thresholds until administration capacity improves. - VAT registration should require proof of identity and basic trading information; refuse registration where preregistration checks indicate strong fraud risk; financial security may be required where doubts remain. - Establish forecasting and monitoring systems to anticipate refund levels and ensure funds availability. - Process refunds (paid, offset, or denied) within a reasonable statutory period (e.g., 30 days). Extensions permitted where returns incomplete, taxpayer has outstanding returns, taxpayer fails to respond to verification, or reasonable suspicion of inaccuracy/fraud exists. - Report publicly on performance in meeting statutory refund deadlines. - Pay interest on late refunds to compensate legitimate claimants deprived of working capital. - Offset excess VAT credits only against VAT and other tax arrears (not nontax debts), except where adequate accounting and debt management systems exist; avoid offsetting anticipated liabilities unless high noncompliance history. - Immediate refunds for exporters or enterprises with at least 50 percent export turnover; other taxpayers may carry forward excess credits for six months, with refund if excess remains. - Verification should be part of a broader audit program; limit pre-refund audits to high-risk cases and use selective post-refund audits otherwise. - Preferential treatment for regular exporters with sound compliance histories (approved refund levels, claimant categorization). - Apply appropriate sanctions and prosecute refund-related fraud through criminal justice. - Ensure appeal rights to independent tribunals for withheld refunds. - Provide clear taxpayer information and simple refund claim procedures. ### XVII. Conclusions (excerpted) - Tax administrations risk abandoning self-assessment principles when addressing refund problems, introducing intrusive, costly measures that apply to all taxpayers regardless of compliance behavior. - VAT is feasible only as a self-assessed tax; substitutes for effective risk-based approaches are unlikely to provide sustainable solutions to refund-related compliance problems. *IMF Working Paper — Appendix I (survey-based analysis and tables).* ### Appendix I ............................................................................................................. ### Appendix I ### I. Introduction - IMF’s Fiscal Affairs Department (FAD) has provided substantial technical assistance in implementing and improving value-added tax (VAT) systems in developing and transitional countries over the past two decades. - The VAT is now a key component of the tax system in over 130 countries at different stages of economic development, raising about 25 percent of the world’s tax revenue. - This working paper follows up on refunding VAT excess credits and is based on responses to a survey of tax administrations in 36 developing, transitional, and developed countries. - The survey requested information in four key areas: - (1) general information on the VAT system operating in each country (e.g., registration threshold, VAT rates, items subject to zero-rating, and number of VAT payers); - (2) details of each country’s VAT refund system (e.g., statutory provisions relevant to the treatment of excess VAT credits, categories of refund recipients, number and size of refund claims, procedures followed by refund claimants, organizational arrangements for processing VAT refunds, and specific regimes introduced to counter administrative problems); - (3) specific data relating to tax audits and identification of cases of refund fraud; and - (4) evaluation and comments on the efficiency and effectiveness of the VAT refund system, including proposals for improvement. ### II. Overview — key features and problems - Invoice-credit VAT feature: - Some businesses pay more VAT on purchases than they collect on taxable sales and should reclaim the difference; particularly true of exporters whose export sales are zero-rated. - Magnitude and regional patterns: - In many countries, VAT refund levels exceed 40 percent of gross VAT collections. - Forty percent of survey respondents repay a third or more of gross VAT collections in refunds. - Countries with refund levels below 20 percent are mostly in Africa, Asia, and Latin America. - Timeliness and administrative practice: - In most developed countries refunds generally paid within four weeks of a refund claim. - In developing and transitional countries it often takes several months, and sometimes more than a year, to process refund claims. - Delays undermine exporter competitiveness and working capital for businesses. - Causes of delays and safeguards: - Prevalence of fraudulent claims often cited as reason for delaying refunds; less advanced tax administrations perform time-consuming verification checks leading to backlogs. - Delays also occur when state budgets are under pressure and when tax collection targets are not met. - Administrations with forecasting and budgeting capabilities can predict refund levels with fair precision. - Legal frameworks and remedies: - 90 percent of IMF survey respondents reported tax authorities are bound by law to making refunds within a prescribed timeframe, generally 30 days. - Around 40 percent of surveyed countries provide for interest to be paid on late refunds. - In 60 percent of surveyed countries, mandatory carry-forward periods for excess VAT credits are imposed, generally for nonexporters. - Alternative arrangements and their trade-offs: - Zero-rating supplies to exporters (used by some EU countries, North Africa, and Asia) and deferral or exemption on VAT owing on imported capital goods are used to reduce refund claims but add administrative complexity and revenue risks. - Large-scale cross-checking of purchases and sales (e.g., Azerbaijan, Bulgaria, China, Korea) and VAT bank account schemes (e.g., Bulgaria) lock away portion of enterprises’ working capital but increase compliance costs. - Some countries deny VAT credits for large purchases paid in cash (credit only when payment made through banks). - Kenya requires all large refund claims to be certified by registered CPAs, effectively outsourcing refund verification to accounting professionals. - Growing trend to introduce fast-track refund processing for taxpayers with proven records of good compliance. ### III. Country experiences with VAT refunds — summary findings - Key survey findings: - Refunds can be substantial, both in absolute terms and as a percentage of VAT collections. - Refund levels vary widely from region to region. - Refund levels are typically higher in advanced and emerging economies. - Within regions, refund levels are largely similar among countries with similar VAT systems and economic conditions. - A pattern of refund claims tends to develop within countries over time, with refund levels relatively constant from year to year. - Refund claims are dominated by exporters. - Most countries have statutory deadlines for refunds, but these are often not met in practice. - More than half the countries surveyed do not provide for interest to be paid on late refunds. - All countries report VAT refund abuse, but most have difficulty estimating the scale of associated revenue losses. - VAT refund abuse is only one component of VAT fraud; some administrations focus audit resources mainly on refunds and neglect other VAT fraud/evasion risks. ### IV. Size of refund claims — regional and country statistics - Table 1. Value of VAT Refunds by Country/Region (In percent of gross VAT collections) — Average1 - Canada 50.3 - EU 38.1 - Eastern Europe 36.8 - New Zealand 35.5 - Former Soviet Union countries 29.6 - Latin America 17.4 - Middle East 16.2 - Asia (not including Singapore) 7.0 - Africa (not including South Africa) 6.0 - 1 Average refund level over a four-year period (1998 to 2001). - Table 2. Value of VAT Refunds in Advanced, Transitional, and Emerging Economies (In percent of gross VAT collections) — Average1 - Advanced Economies - Canada 50.3 - France 21.2 - Ireland 24.9 - Netherlands 50.0 - New Zealand 35.5 - Sweden 48.6 - United Kingdom 40.9 - Transitional Economies - Bulgaria 21.5 - Hungary 48.2 - Latvia 49.1 - Romania 24.7 - Russia 44.6 - Slovak Rep. 53.9 - Ukraine 24.1 - Emerging Economies - Chile 28.8 - Colombia 4.1 - Indonesia 12.4 - Mexico 32.1 - Morocco 5.1 - South Africa 39.5 - Others - Algeria 24.3 - Bolivia 10.4 - Cambodia 2.8 - Cameroon 8.8 - El Salvador 9.6 - Kenya 7.2 - Mozambique 2.7 - Peru 19.8 - 1 Average refund level over a four-year period (1998 to 2001). ### V. Determinants of refund levels and a simple identity - Factors influencing a country’s VAT refund level (percent of gross VAT collections): - (1) Nature of the economy (extent to which investment generates excess VAT credits, value-added of export industries, proportion of taxable and zero-rated sales). - (2) Design of the VAT system (extent of zero-rating and use of multiple rates). - (3) Taxpayer compliance behavior and extent of VAT fraud. - (4) System and culture of the tax administration (level of corruption, capacity to detect and prevent VAT fraud, commitment to taxpayer service). - Simple identity for refunds under a fully functioning, single-rate VAT: - (α I + β(1 – λ)Z)/ẽ - Definitions: - I and Z denote the shares of investment and zero-rated items (including exports) in GDP; - α is the proportion of investment that generates excess credits; - β is the proportion of zero-rated sales that generates excess credits; - λ is the ratio of value added to sales in the zero-rated sector; - ẽ is the gross efficiency ratio (gross VAT collections as a percent of GDP per percentage point of tax). - Example calculation using the identity: - Suppose investment ratio I = 10 percent, exports Z = 40 percent of GDP, α = 5 percent (5 percent of investment generates excess credits), value-added λ = 40 percent of sales in export sector, β = 1 (exporters do not supply significant domestic market), and ẽ = 0.9 (gross efficiency ratio). - Then refunds equal 27 percent of gross collections. - Empirical modeling (Box 1 summary): - Empirical estimates predict variations in refund-to-collections ratios under different institutional arrangements: - Example prediction: for a country with high literacy, exports equal to 30 percent of GDP, GDP growth 3 percent, and a single nonzero VAT rate: - Predicted refunds would constitute about 37 percent of gross collections if refunds are paid through budget appropriation. - Predicted refunds would constitute about 41 percent of gross collections if refunds are paid out of revenue. - If supplies to exporters are zero-rated (D1 = 1), these ratios fall to about 12 percent and 16 percent respectively. - If country exhibits characteristics similar to those in the last column of Table 2 (D3 = 1), ratios fall to about 20 percent and 24 percent respectively. ### VI. Administrative responses and performance considerations - Tax administrations adopt a range of measures to manage refund risk and timeliness: - Limit pre-refund verification to perceived high-risk claims within a risk-management compliance strategy. - Use statutory powers to conduct audits and verification checks, require security or bank guarantees. - Apply zero-rating to exporters or defer VAT on imported capital goods to reduce refund claims (trade-off: added complexity and revenue risk). - Cross-check purchases and sales transaction data at large scale (examples cited: Korea, China, Bulgaria). - Require VAT payments through banks to deny input credits on large cash purchases (used where large shadow economies exist). - Outsource verification functions (e.g., Kenya’s requirement for CPA certification of large refund claims). - Fast-track refunds for taxpayers with proven compliance records. *IMF Working Paper — Appendix I (survey-based analysis and tables).* ### Box 1. The Empirical Modeling of Refunds ### Box 1. The Empirical Modeling of Refunds ### A. Empirical model and key regression results - Estimated equation: - Refunds = 0.16 Exports + .75 Growth + .19 Literacy + .90 Range – 25.3 D1 + 3.8 D2 – 17.5 D3 - (2.06)* (0.69) (3.41)* (2.6)* (-2.51)* (0.83) (-3.70)* - t-ratios reported in parentheses; an asterisk denotes significance at 5 percent. - Adjusted R-squared = 0.8826. - Variable definitions: - Refunds = average of refunds paid to gross VAT collections over the survey period 1998-2001. - Exports = share of exports in GDP (in percent). - Growth = average GDP growth rate over the survey period (in percent). - Literacy = literacy rate (in percent). - Range = difference between the highest and lowest (nonzero) VAT rates (in percent). - D1 = dummy = 1 if supplies to exporters are zero-rated. - D2 = dummy = 1 when refunds are paid from gross collections (rather than as an expenditure appropriation). - D3 = dummy = 1 for the “other” economies in Table 2 (developing countries that are neither transitional nor emerging and are outliers with generally weak refund performance). - Main empirical findings from the equation: - Refunds rise with openness (Exports) and this variable is significant at the 5 percent level. - Refunds fall with less mature tax administrations (D3), significant at the 5 percent level. - Growth has a positive effect on the refund ratio, but the coefficient is not significant. - Literacy has a positive and significant effect, capturing development of the economy. - Range of VAT rates has a positive and significant effect, capturing refunds generated on domestic sales at preferential rates. - Zero-rating supplies to exporters (D1) lowers the refund ratio, as expected (D1 coefficient negative and significant). - Paying refunds out of general revenue (D2) has a positive effect on refunds (coefficient positive, not significant in this equation), with the positive sign robust across specifications; implies explicit budgetary appropriations for VAT refunds may retard refund payments. ### B. Refund recipients and composition of claims - Exporters dominate VAT refund claims in both number and value. - Typical institutional and claimant patterns: - Many VAT laws limit entitlement to refunds to exporters; nonexporting enterprises often required to carry forward excess credits to subsequent tax periods. - A small number of large exporters typically account for the majority of VAT excess credits refunded. - Exporters’ share of total VAT refund claims, 2001 (selected country examples; shares are "in percent of total claims"): - Cameroon: Share in Number of Claims to Total Claims = 60; Share in Value of Claims to Total Claims = 66. - Kenya: Share in Number of Claims to Total Claims = 70; Share in Value of Claims to Total Claims = 48. - Morocco: Share in Number of Claims to Total Claims = 80; Share in Value of Claims to Total Claims = 80. - Cambodia: Share in Number of Claims to Total Claims = 76; Share in Value of Claims to Total Claims = 41. - Slovak Republic: Share in Number of Claims to Total Claims = 56; Share in Value of Claims to Total Claims = 63. - Kazakhstan: Share in Number of Claims to Total Claims = 100; Share in Value of Claims to Total Claims = 100. - Bolivia: Share in Number of Claims to Total Claims = 100; Share in Value of Claims to Total Claims = 100. - Chile: Share in Number of Claims to Total Claims = 64; Share in Value of Claims to Total Claims = 89. - Colombia: Share in Number of Claims to Total Claims = 53; Share in Value of Claims to Total Claims = 63. - El Salvador: Share in Number of Claims to Total Claims = 100; Share in Value of Claims to Total Claims = 100. - Peru: Share in Number of Claims to Total Claims = 100; Share in Value of Claims to Total Claims = 100. - Source: IMF survey responses. - Other common categories of nonexporter claimants: - Registered taxpayers supplying zero-rated goods and services to the domestic market (example given: hospitals and universities in Australia). - Registered traders with excess credits from temporary trading conditions (seasonal slump). - Registered entities with large purchases of capital goods relative to current sales (start-ups, replacement machinery). - Registered traders subject to a dual rate structure (refunds arise where outputs taxed at reduced rate and inputs taxed at higher standard rate); limited data, example: Slovak Republic ~7 percent by number and 15 percent by value of total claims in 2001. - Registered traders subject to withholding arrangements (noted prevalence in Latin America and West Africa; risk of proliferating refund claims if withholding rates set too high). - Claimants not registered for VAT: (1) diplomats and bodies exempted under diplomatic conventions; (2) visiting tourists entitled to refunds on goods taken home (refunds may be limited to large purchases within a specified time before departure). ### C. Time taken to process refund claims, statutory rules, and administrative practices - Statutory deadlines and prevalence: - 90 percent of survey respondents report tax authorities are required to make refunds within a prescribed timeframe. - Statutory refund timeframes in the sample range from 24 hours (Peru, where security is provided by the claimant) to 90 days (France). - Most common statutory period = 30 days (40 percent of the survey countries have a 30-day refund period). - French VAT code stipulates 90 days, but authorities apply an administrative performance standard of 30 days. - Rationale for statutory deadlines: - Make VAT operation fair by aligning timeframes for paying and refunding VAT. - Help reduce corrupt practices by limiting opportunities for officials to extract payments to speed refunds. - Statutory deadlines often not met: - Shortcomings due to weaknesses in refund processing systems or government short-term cash shortfalls. - Delays occur in developing, transitional, and advanced economies. - Interest on late refunds: - Around 40 percent of surveyed countries provide for interest on late refunds. - De minimus rules may apply (examples: Singapore and the United Kingdom—no interest if calculation less than prescribed statutory amount). - Interest calculated at a statutory rate often aligned to prevailing commercial bank interest rates and adjusted quarterly or half-yearly by regulation, multiplied by days elapsed since statutory deadline expiration. - Carry-forward rules: - 60 percent of surveyed countries require taxpayers, particularly nonexporters, to carry forward excess VAT credits for a specified period; refund paid only if excess remains at end of carry-forward period. - Carry-forward periods range from 30 days to more than a year; generally in the range of three to six months. - EU’s Sixth Directive: Member States may either make a refund or carry the excess forward to the following period according to conditions they determine. - France: norm is carry-forward for nonexporters with refunding as the exception. - Ireland, the United Kingdom, Sweden, and the Netherlands reported no mandatory carry-forward periods. - Carry-forward measures are burdensome for firms making large preoperational capital investments; example policy response: Albania allows businesses to defer VAT liabilities on certain imported capital goods. - Offsetting refunds against other tax liabilities: - VAT laws in 80 percent of surveyed countries allow tax authorities to offset VAT refunds against other tax debts (e.g., income tax). - Conditions vary: some require tax liability to be due and payable before offset; others allow offset even if amounts are not yet due. - In some former Soviet Union states, offsetting may extend to nontax debts owing to the state. - Examples: Canada and Hungary allow offsetting against customs duties as well as other tax liabilities. - Discretion vs mandatory offsetting: Australia provides limited discretion—refund rather than offset in specific situations (tax debt not yet payable; tax debt subject to installment arrangements; recovery deferred because assessment in dispute). - FAD technical advice on offsetting (qualified): - VAT refunds should be offset only against other tax liabilities (not nontax debts). - Generally should not offset against anticipated tax liabilities (taxes assessed but not yet due) because of negative cash-flow effects on taxpayers; exception for taxpayers with a history of noncompliance. - Offsetting should be adopted only if the tax authority has an adequate taxpayer accounting system and debt management infrastructure. - Modern unified tax administrations implement integrated computerized accounting systems for consolidated views of taxpayer liabilities and entitlements. - Administrative thresholds and extraordinary measures: - Some countries (Italy, France, Peru) reported not processing refund claims below a specified threshold. - At least two African countries enacted legislation denying payment of all refund claims outstanding as of a specified date to wipe large backlogs—practice noted as having serious long-term consequences for tax system integrity and credibility of tax administration. ### D. VAT refund abuse and estimates of revenue loss - VAT refund abuse leads to substantial VAT revenue loss; causes include error, deliberate understatement of VAT liabilities, and systematic attacks on refund systems. - U.K. estimates (2002–03 context referenced): - Authorities estimated VAT losses to be around 15.8 percent of net VAT receipts in 2002–03 (made up of fraud and nonfraud losses). - In 2002–03, U.K. VAT receipts were £108.8 billion, of which £45.2 billion was refunded (source: Tackling VAT Fraud, report of the U.K. National Audit Office, March 2004). - Table 4. Estimate of VAT Revenue Losses in the United Kingdom in 2001–02 (area and estimate in £ billion): - Noncompliance by traders (genuine mistakes or deliberate understatement/inflation of purchases): 2.5 to 4.0. - VAT “missing trader” fraud (fraudsters register for VAT, buy VAT-free from another EU member state, sell at VAT-inclusive prices, then disappear): 1.77 to 2.75. - Traders failing to register where turnover exceeds threshold: 0.4 to 0.5. - VAT avoidance schemes that businesses purport to be legal but are (or are likely to be) challenged by the tax authority: estimate not provided in supplied excerpt. *Source: _wp05218 - Box 1. The Empirical Modeling of Refunds* ### 2.5 to 3.0 ### _wp05218 - 2.5 to 3.0 ### Context and headline figures - Table excerpts: "2.5 to 3.0" and "Total 7.17 to 10.25". - During 2002–03, U.K. authorities disallowed "£63 million" of VAT refunds to exporters suspected of being part of a supply chain where VAT had gone missing through a “carousel” fraud. - The U.K. tax authority target: stop long-term growth in the VAT gap and cut it from "15.8 percent" in "2002–03" to "12 percent" of the total amount that could be theoretically collected from VAT by "2005–06". ### Nature of VAT fraud, risk management uptake, and verification practice - Fraud and evasion methods are similar across countries; policy responses differ: - Risk-management grounded methods used in: Hungary, New Zealand, the United Kingdom. - Intrusive systems applied to all taxpayers used in: Azerbaijan, Bulgaria, China, Korea. - Two-thirds of surveyed countries reported they undertake risk-assessments in processing VAT refund claims. - Risk-assessment scope varies from highly developed systems (e.g., United Kingdom: wide-ranging data, computer applications, statistical methods to identify suspicious transactions) to rudimentary subjective processes without computer support. - Contradictory survey responses: several African and Latin American respondents report both conducting risk-assessments and verifying every VAT refund claim prior to payment. - A quarter of survey respondents indicated a statutory requirement to verify every refund claim prior to payment. - Some countries without statutory requirements verify all claims as an administrative practice. - Example: Tanzania — under VAT laws every refund claim must be verified by an auditor registered by the Tanzanian national board of accountants and auditors. - A third of countries reported they do not have a VAT audit program. - For countries that do have audit programs, pre-refund audits dominate in a quarter of countries. ### Reasons cited for weak audit programs (as reported by many developing and transitional countries) - (1) Insufficient numbers of highly skilled and appropriately remunerated audit practitioners; - (2) Authorities’ concerns about collusion between taxpayers and auditors; - (3) Inadequate preparation at the time of VAT implementation, possibly because the consequences of a weak audit program were not immediately perceptible; - (4) Lack of clear political support for the tax administration; - (5) Lack of an appropriate legal and judicial environment. ### Information exchange and inter-agency cooperation - Exchange of VAT and income tax information: - Where direct and indirect taxes are administered together in a unified tax administration, exchange normally happens automatically. - In countries where VAT and income tax are administered by different agencies, exchange is often limited; example: in some countries (e.g., Egypt and Tanzania) income tax law does not allow income tax-related information to be provided to the VAT administration. - Footnote: "Currently, of around 140 countries that have adopted a VAT, 120 or so have integrated function-based revenue administrations covering both income tax and VAT—see The Modern VAT." - Exchange of VAT and customs information: - All surveyed countries have some form of information exchange between tax and customs agencies for verifying VAT returns and refund claims. - Information flow ranges from case-by-case requests (e.g., Mozambique and Tanzania) to periodic routine data transmission (monthly import/export data) and, in some cases, on-line customs database access (e.g., Singapore). - Five countries where VAT and customs administration were carried out by the same government entity at the time of the survey: Canada, Mexico, Peru, South Africa, and the United Kingdom. - Exchange of information among countries: - EU example: member states exchange information about VAT-registered traders and intra-Community supplies via the EU’s VAT Information and Exchange System (VIES). - U.K. bilateral agreements with eight member states (with another three in the pipeline) to allow more rapid information exchange to help identify VAT fraudsters. ### Budgeting and organizational arrangements for VAT refunds - Less than "40 percent" of surveyed countries make specific expenditure appropriation for VAT refunds in their annual budgets. - Most respondents pay refunds out of consolidated VAT revenue collections. - Two budgeting methods for VAT refunds: - Pay from gross VAT revenue (used by "60 percent" of survey respondents). - Pay from budget expenditure appropriations. - Organizational arrangements: - "47 percent" of survey countries have dedicated organizational units responsible for processing VAT refunds. - Other models: refunds processed as part of broader returns-processing operations (e.g., Sweden, Hungary, Slovak Republic) or audit operations (e.g., Cambodia). - Some countries use central automated processing centers (e.g., Netherlands, New Zealand, United Kingdom). - Countries unwilling to embrace self-assessment are more inclined to establish dedicated VAT refund units to control transactions and verify all refund claims, preferably prior to payment. ### Administering VAT refunds in a self-assessment environment - Key principles: - VAT operates so businesses that pay more VAT on purchases than they collect can reclaim the difference; refunds are essential to avoid making VAT a tax on production. - Modern VAT systems rely on voluntary compliance and self-assessment, with taxpayers calculating liabilities and refund entitlements, filing returns, and claiming refunds. - Tax administration role: assist taxpayers to understand obligations and entitlements, and take action against non-compliers, focusing on highest revenue risks. - Administrations need clear laws and procedures, adequate resources, compliance programs mixing education, assistance, enforcement, and verification, and enforcement tools (audit, reassessment, collection, penalties). ### VAT compliance strategy and targets - Compliance strategy design: - Apply risk-management principles and balance revenue protection with not overburdening compliant taxpayers. - A well-developed program identifies major risks, responses, and criteria for evaluating progress. - Example U.K. strategy elements to reduce VAT gap: - Making it as easy as possible for legitimate traders to pay their VAT; - Making it as difficult as possible for dishonest traders to commit VAT fraud; - Challenging, through the courts, abusive VAT avoidance schemes; - Detecting unregistered trading; - Identifying traders who do not pay the correct amount of VAT. ### Detecting, investigating, and preventing fraud — practices and audit design features - Fraud and evasion vary from occasional omission of a sale to systematic suppression of sales and falsified invoices; some fraudsters register solely to steal VAT via refunds. - Estimating VAT losses from refund abuse is difficult; requires specialist skills, statistical and economic assumptions, and interpretation of uncertain data; some administrations use economic consultants and academic institutions. - Intelligence and risk-assessment: - Information gathering and intelligence work identify high-risk sectors and trader profiles and target organized crime networks. - Advanced administrations (e.g., U.K.) gather intelligence from customs, inland revenue, other government agencies, and use computer systems and statistical methods to compare VAT returns with sector trends. - Effective audit program design features and principles: - A broad coverage of taxpayer groups, by size and by sector, and of compliance issues; - Audit resources spread across all elements of the program, avoiding disproportionate absorption in verifying refund claims prior to payment; - Pre-refund audits limited to high-risk cases only (e.g., first refund claim by a new registrant or claims that vary significantly from established patterns), with lower-risk claims subject to selective post-refund audits; - VAT audits that are primarily short, issue-oriented, and limited to one or two tax periods; - Audits focused on accounting systems rather than individual transactions, especially for large taxpayers; - Close coordination of VAT audit program with audit programs of other taxes, particularly income tax, including comprehensive audits covering all a taxpayer’s obligations over multiple periods; - Consistent application of appropriate penalties for noncompliance; - Investigation of serious fraud cases with a view to prosecution under the criminal code. ### Box 2 — Types of VAT fraud and evasion (summary) - Inflated refund claims: creating fake invoices for purchases never made; organized crime may fabricate businesses to generate invoices. - Underreported sales: most common evasion method; conceal domestic sales to evade VAT and generate excess credits to be refunded. - Fictitious traders: short-lived sham enterprises that register for VAT and invent fake export invoices to claim refunds. - Domestic sales disguised as exports: sales on the domestic market claimed as exports using fake export invoices. - Missing trader intra-EU fraud: a trader registers in an EU country, purchases goods VAT-free from another member state, sells at VAT-inclusive prices, then disappears without remitting VAT. - Carousel fraud variant: goods are sold through contrived transactions and sold back to the originating country, allowing repeated fraud using the same goods. - Other forms: - Traders liable to VAT but not registering; - Credit claimed for taxable supplies used in exempt activities, and credit claimed on private purchases; - Credit claimed for invoices from unregistered suppliers; - Illegally imported goods sold with VAT added but not remitted; and - Barter arrangements hidden from authorities. *Italicized source attribution: IMF staff paper excerpt, _wp05218 - 2.5 to 3.0.* ### Box 3 describes the range of audit types undertaken by modern tax administrations, ### _wp05218 - Box 3 describes the range of audit types undertaken by modern tax administrations, ### Range of audit types - In a function-based tax administration, the audit program typically includes the following range of audits: - Registration checks: quick checks to establish whether businesses are correctly registered for all tax obligations, including VAT; information from third parties (e.g., customs and business licensing centers) and other audit activities may trigger checks. - Advisory audits: auditors visit new enterprises to advise on obligations and entitlements (filing and payment, refund claims, record-keeping, risk of audit, sanctions); particularly appropriate when introducing new tax laws. - Record-keeping audits: unannounced visits to business premises to check that appropriate records are kept and that VAT invoices are being issued. - Desk audits: basic checks from the tax office (examining VAT and income tax returns, selective cross-checking, ratio analysis); generally apply to specific areas of enquiry, small business enterprises and individual (non-business) taxpayers; may lead to further investigation. - Single-issue audits: focus on a single tax type, or a specific (usually limited) tax period; for VAT this would cover one or two returns only. - VAT refund audits: verify a taxpayer’s entitlement to a refund prior to processing; usually undertaken for a taxpayer’s first refund claim and where a refund claim varies significantly from established patterns and trends; may include outbound telephone verification and requests for copies of invoices to substantiate claims. - Audit projects: managed on a project basis covering a specific group of taxpayers (industry or line of business) and/or certain return items; involve specific checks designed to address particular risks or determine compliance levels in a sector. - Comprehensive (or full) audits: cover all tax obligations over a number of tax periods; typically undertaken after discovery of discrepancies during single-issue audits; time-consuming and comprehensive; applied only to taxpayers showing evidence of underreporting across income tax, VAT, and other taxes. - Fraud investigations: involve the most serious cases with criminal implications; require special investigation skills, meeting evidentiary requirements, often involve seizure of records, taking testimony, and preparing briefs for courts. - Verification of refund claims is one component of a much wider risk-based audit program aimed at broad coverage of taxpayers and compliance issues. ### VAT registration (Section E) - A tax authority’s ability to administer the VAT effectively—including the processing of refund claims—is influenced by the level of the threshold for compulsory registration. - Experience suggests many countries have tended to set the threshold too low, creating difficulties when tax administration capacity is insufficient to manage the number of registered taxpayers. - Recommendation frequently made by FAD: - Regulate the number of VAT taxpayers (by adjusting the VAT registration threshold) at a level that can be realistically managed by the tax administration. - Where administration is weak, maintain a high VAT threshold until the tax authority’s capacities are developed to enable it to administer a larger number of VAT taxpayers in a self-assessment environment (i.e., until the tax authority is organized appropriately, has adequate resources and systems, and has established effective enforcement and service programs). - VAT registration process controls to prevent registration of fictitious traders: - Applicants should provide proof of identity and basic information about intended trading activities (nature and location of business operations; anticipated turnover; type of goods or services; sources of supply; sources of business finance). - Registration staff should assess this information against risk criteria; a short interview may be necessary; automatic computer checks of internal and external databases may be undertaken. - For the vast majority of cases, these procedures should establish authenticity without significant cost or delay. - Where registration officers have doubts, refer to enforcement staff for further checking, including a visit to business premises and verification of sources of business finance. - VAT registration should be refused where preregistration checks establish a strong risk of fraud. - If preregistration checks raise suspicions but evidence is insufficient to refuse registration, the tax authority may require a financial security to be lodged by the applicant (including demonstration of the source of the financial security paid). ### Taxpayer service (Section F) - Growing trend to introduce fast-track refund processing for taxpayers with proven records of good compliance in response to business community demands. - Governments are increasingly sensitive to business needs and costs associated with VAT systems; exporters particularly complain about cash-flow impacts of delayed refunds. - Tax administrations are increasingly emphasizing helping traders understand VAT obligations and entitlements, including requirements relating to refund claims. - Priority assistance often given to new businesses (in cooperation with industry bodies and other government agencies) and taxpayers operating in the shadow economy who may not be fully aware of VAT registration requirements. - Close attention to new registrants can also identify and prevent potential fictitious traders from committing fraud through the VAT refund system. - Approaches that give preferential treatment to good compliers are discussed in Section V. ### Specific approaches to VAT refund problems (beginning of Section V) - This section describes and evaluates specific approaches adopted by several countries to deal with VAT refund problems, sometimes as substitutes for an effective VAT compliance strategy based on risk-management principles. A. Zero-Rating Supplies to Exporters - Historical and practical notes: - France implemented a scheme (Le système des achats en franchises) when adopting the first VAT-type tax in 1948 to allow regular exporters to purchase inputs free of tax, effectively zero-rating supplies to exporters to eliminate the need for exporters to claim refunds of excess credits; scheme further developed in 1954 and 1968. - Some other countries (e.g., Ireland, Italy, the Netherlands) adopted zero-rating for exporters when VAT was introduced in the 1970s; the Netherlands later abandoned it in the 1990s. - The Sixth Directive provides EU member states the option to adopt this system, but only a limited number have done so. - The scheme was introduced in several former French colonies in the 1980s (including Algeria, Côte d’Ivoire, Morocco, Tunisia, and Senegal) and in the 1990s in a few transition and emerging market countries (e.g., Korea, Albania, and Azerbaijan). - Administrative observations and risks: - The scheme adds complexity and revenue risks to VAT administration because it breaks the VAT credit chain. - Zero-rating supplies to exporters shifts the control problem from a small number of well-known exporters to a larger and lesser-known group of suppliers. - Certificate mechanisms (e.g., Azerbaijan) are open to abuse and add to administrative workloads in monitoring downstream suppliers. - Footnote summaries from source: - Current French scheme limited to direct exporters and operates under self-assessment principles; exporters advise suppliers not to charge VAT except for capital goods; zero-rated supplies subject to an annual ceiling equal to total value of exports in the previous year; year-end reporting of purchases and exports; new exporters must seek approval; delinquent taxpayers can be removed. - Ireland: traders who export more than 75 percent of their output can obtain an authorization that allows their suppliers not to charge VAT. - Azerbaijan: zero-rate authorization certificates issued to exporters in the hydrocarbon sector and their direct suppliers; certificates passed down the supplier chain. - Modernization of French tax administration (integration of direct and indirect tax administration from 1948; development of comprehensive audit programs from 1954; establishment of full function-based tax offices from 1968) is closely linked to implementation and improvement of its VAT system. B. Large-Scale Cross-Checking of Invoices - Some countries have attempted computerized cross-checking of all purchases and sales invoices to validate VAT credit claims and identify undisclosed sales. - Korea commenced an ambitious cross-checking program in the late 1970s. - More recent attempts made by China and a handful of other countries (e.g., Indonesia, Bulgaria, Azerbaijan, and Albania). - Typical system feature: taxpayers submit copies (or a list) of invoices with regular VAT returns; details are entered into a central database. - Box 4 in the source briefly describes the experiences of Korea and China. *Source: _wp05218 - Box 3 describes the range of audit types undertaken by modern tax administrations,* ### Box 4. Large-Scale Cross-Checking of Invoices—Korea and China ### Box 4. Large-Scale Cross-Checking of Invoices—Korea and China ### Korea: early invoice-matching system and current practice - Development began in the 1970s with a computer-based invoice-matching system when electronic data capture was not readily available. - System required the tax administration to transcribe data from paper copies of invoices supplied by vendors and purchasers. - Data capture was time-consuming, costly, and prone to error. - Taxpayer compliance costs were high because suppliers and purchasers were required to submit copies of invoices with their VAT returns. - The system identified numerous mismatches; considerable administrative resources were consumed examining these. - Most mismatches were not due to fraudulent claims, but were the result of transcription errors, incomplete data, and valid timing differences. - Current practice: general VAT taxpayers supply the tax authority with summary information of purchases and sales invoices (replacing earlier requirement to submit copies of invoices). ### China: Golden Tax Program (GTP) - China developed a computerized cross-checking system called the “Golden Tax Program” (GTP) to: - identify false invoices, - verify credit entitlements, - check reported sales on VAT returns, - assist audit case selection generally. - Coverage and application: - The system does not operate in all provinces. - It is applied to mid-scale and large enterprises. - Operational details: - A VAT taxpayer is given a smartcard and software (the “black box”), activated when an invoice for a taxable supply is generated within the taxpayer’s computerized accounting system. - The “black box” generates an encrypted code—an 84-digit number—which is printed on the VAT invoice (VAT invoice forms are accountable documents supplied by the authorities). - Information contained in the encrypted code includes the registration numbers of the supplier and purchaser, the sales price, the nature of the goods, and VAT charged. - This information is stored on an electronic file that, at the end of the tax period, is copied to a disk and submitted to the tax authorities for input to the GTP database. - Invoice details supplied by purchasers (when they file their VAT returns) are subsequently cross-checked with the information held in the database. - Reported outcomes and unknowns: - Authorities report a positive impact on VAT compliance, but no supporting details were obtained. - Details of the costs of developing and administering the system, and taxpayer compliance costs, are unknown. ### Diagnostic findings on large-scale invoice cross-checking systems - Despite improvements in information technology and automated data capture, administrative and taxpayer compliance burdens associated with large-scale invoice matching remain significant. - Recent FAD diagnostic reviews found these systems continue to generate considerable unproductive work. - Example observation: in one country, more than half of the large taxpayer unit’s highly skilled auditors were substantially engaged in following-up and checking invoice discrepancies reported from the invoice matching system. - Many discrepancies related to data entry errors (e.g., errors in taxpayer identification numbers and addresses), creating administrative costs with no associated revenue benefits. - FAD assessment and recommendation: - Large-scale cross-checking systems are a poor substitute for well-designed audit programs based on risk assessments, selective cross-checking, intelligence gathering, and targeted fraud investigation. - The net benefits of large-scale cross-checking systems are yet to be proven; associated costs to businesses and tax administrations remain unacceptably high. - Cross-checking should be directed at industries and taxpayer groups exhibiting the highest potential for invoice-related fraud, and should be applied on a sample basis or where a tax auditor has grounds for suspicion. ### C. Certification of Refund Claims by CPAs (Kenya example) - Legal requirement: - Under Kenya’s VAT laws, refund claims exceeding a specified amount must be certified by a CPA registered with the Institute of Certified Public Accountants of Kenya. - The law imposes sanctions on accountants who knowingly certify false claims. - Reported benefits and impacts: - Tax officials argue it helps eliminate fraudulent claims and reduces administrative costs. - Kenyan officials report that the number of refund claims by exporters dropped by 40 percent following introduction of the scheme, suggesting many firms had been submitting false claims. - Large exporters support the arrangement because it speeds up refunds; they are willing to bear increased compliance costs. - CPA firms favor it because of the opportunity to generate service fees. - Design considerations for effectiveness: - Certification required for larger refund claims only (statutory threshold set at an appropriate level) to avoid burdening small refund requesters. - Requires a high-integrity (noncorrupt) accounting profession, a sufficiently strong tax administration, and a sound judicial system to enforce sanctions. - Requires acceptance by traders of additional compliance costs. ### D. Preferential Treatment of Good Compliers (“Gold” Status Scheme) - Concept: - Preferential treatment for taxpayers with sound compliance histories (especially exporters), e.g., “gold” status for accelerated VAT refunds; “silver” and lower groups receive less prompt treatment. - Frees up scarce audit resources for more productive work. - Implementation example: Pakistan (late 1990s and later improvements) - Initially manual procedures; later basic computer applications developed to provide information on traders’ compliance history using VAT, income tax, and customs data. - Refund claimant categories: 1. “Gold” — minimal revenue risk. 2. “Silver” — moderate risk. 3. “Others” — high or unknown risk. - Processing times and controls: - Gold refund claimants normally have their claims approved for payment without a pre-refund audit within 3-5 days. - Silver claimants are assigned an upper refund limit; claims not exceeding the limit are subject to a brief desk review and approval within 15 days. - Post-refund audits are conducted at least once a year on two or three claims submitted in the past 12 months by gold and silver claimants. - If post-refund audits detect persistent inaccurate claims, gold or silver status is withdrawn. - Criteria used to qualify as “Gold” and “Silver” (selected elements): - Gold: Exporters with at least three years’ export history and net wealth exceeding a specified amount; proper books of account for at least the last three years; no evidence of fraud or significantly overstated refund claims in the past three years; history of accurate and timely tax remittance for all taxes and duties; bank certification that accounts are in good standing; records audited by the tax office for six of the past 24 months. - Silver: Exporters with at least one year export history and net wealth exceeding a specified amount; proper books of account for the duration of taxable activities (or three years, whichever is the lesser); consistent pattern of export activities and products; history of accurate and timely tax remittance for at least the past 12 months; no evidence of fraud or overstated credit claims in at least the past four refund claims; records audited by the tax office for three of the past 24 months. - Selection for pre-refund verification (circumstances): - The claim is a first-time refund claim. - The claim exceeds a value prescribed by the tax administration. - The claim deviates from the regular refund pattern of the claimant. - Previous claims have been rejected or reduced as a result of verification checks. - The claimant has a record of poor compliance in relation to VAT and other taxes (e.g., nonfiling, and late payment). - Program dependence: - The program depends heavily on developing profiles for each claimant, including VAT compliance record, compliance history, and audit results for all other taxes. ### E. Payment for Large Purchases Through the Banking System - Rationale: - Widespread cash transactions facilitate VAT and other tax evasion due to lack of an audit trail. - Tax administrations encourage or force traders to conduct business through the banking system for larger amounts. - Approaches: 1. Transaction threshold rule (used in France, Hungary, Turkey, and others): traders must pay for goods and services above a certain transaction threshold through the banking system; failure to comply results in a financial penalty. 2. Azerbaijan approach: to be entitled to VAT credits on a business-related purchase, a VAT-registered enterprise must make the purchase through the banking system; if the purchase is made in cash, VAT credits are denied. - Expected effects and limitations: - Can reduce unreported transactions by registered traders. - Sometimes less effective than anticipated: - Turkey’s experience: measure did not meet expectations because detection of noncompliance was difficult and the ministerial basis for introduction was challenged in the courts. - Measures are unlikely to impact unregistered traders who remain undetected without programs to bring them into the VAT net. - Note on business disruption: - Forcing bank transactions may be disruptive to business (e.g., concerns over customer solvency, banking system delays). ### F. VAT Bank Accounts (Bulgaria example) - Introduction and objectives: - A VAT bank account system was introduced in Bulgaria in July 2002 to reduce VAT fraud and speed up processing of VAT refund requests. - Under the scheme, each taxpayer registered for VAT must open at least one VAT bank account. - Operational rules: - A purchaser registered as a VAT taxpayer must deposit VAT payments into a seller’s VAT bank account if the VAT charged exceeds a statutorily prescribed threshold. - The VAT deposit must be completed at the time payment for the good or service is made. - VAT input credits will not be denied to a purchaser who has followed the procedures and paid VAT into a seller’s VAT bank account. - Deposit information includes the VAT amount; the taxpayer identification numbers of purchaser and seller; and the VAT invoice number relating to the transaction. - Firms’ VAT bank accounts are used at the close of the VAT period to meet VAT liabilities; funds held in VAT bank accounts can be used only for payment to suppliers of VAT included in prices and for payment of net VAT liabilities due and payable at the end of the tax period. - Monies may be retrieved from VAT bank accounts for other purposes only if: (1) taxpayer seeks permission of the authorities to make the withdrawal; (2) taxpayer undergoes a tax audit; and (3) taxpayer transfers required funds from the VAT bank account to the government bank account pending audit outcome and clearance. - If audit determines other tax liabilities are due, funds will not be returned and will be offset against assessed liabilities. - Administrative details: - Bulgaria’s 80,000 or so VAT-registered businesses must open accounts in banks approved by the tax authority. - The tax authority maintains a list of VAT bank account numbers, by VAT taxpayer name, and the list is publicly available. - Observations on impact and costs: - A full assessment has not been possible due to lack of hard data on impact on VAT revenue and refund delays. - The scheme does not eliminate many types of VAT fraud (e.g., underreporting of sales, false exporting, transactions occurring outside the VAT account system, bribing of tax officials, and false invoicing). - A deposit by a purchaser into a seller’s VAT bank account is no guarantee tax will be deposited into the government’s account. - Compliance costs are considerable: - Loss of working capital—funds held in VAT bank accounts are effectively frozen; businesses may need short-term loans to support cash flow. - Administration costs—businesses incur additional costs in making two payments (exclusive price and VAT) instead of one; number of VAT invoices, payment orders, and bank deposits/withdrawals is doubled; mutual offsets between sellers and purchasers are not allowed; additional account-keeping fees accrue. - Compliance costs are disproportionately higher for small to medium-sized businesses. - Some transitional countries (including Russia) considered the system but decided against it after examining compliance costs. ### G. Deferring Accounting for VAT on Imported Capital Goods - Problem: - Investors importing large capital equipment are required to pay VAT before clearance and then wait for refunds, negatively affecting cash flow and potentially discouraging investment. - Policy alternatives: - Full VAT exemptions for importers (dispenses with refunds) — governments reluctant because of potential abuse and pressure to extend exemptions. - FAD-supported alternative in developing countries: permit VAT taxpayers to defer accounting for the VAT liability on imported capital goods until filing of the next return, allowing importers to offset the VAT liability with the input tax credit to which they are entitled. - Reference: - Box 5 in source describes how the deferral scheme works. *Source: _wp05218 - Box 4. Large-Scale Cross-Checking of Invoices—Korea and China* ### Box 5. Deferral Scheme for VAT Owing on Imported Capital Goods ### Box 5. Deferral Scheme for VAT Owing on Imported Capital Goods ### Scheme Characteristics - The scheme is limited to registered VAT taxpayers who import large items of capital goods. - Capital goods (both imported and domestic) are subject to the standard rate of VAT. - Imports of capital goods by persons who are not registered VAT taxpayers are subject to VAT at the time of import. VAT is paid, as usual, before clearance of the goods. - Importers of capital goods who are registered VAT taxpayers are permitted to defer accounting for the VAT liability until their next return is filed. - In this return, the VAT applicable to those goods is reported as a VAT liability and, in the same return, the VAT input tax credit is claimed for the capital goods. - If the importer is entitled to 100 percent input tax credit (equipment used exclusively in taxable activities) the VAT applicable to the importation, reported as a liability, will be completely offset by the corresponding input tax credit. - The customs office is furnished with a copy of the VAT return to close their records of the importation. ### Evaluation Against Objectives The specific approaches aim to satisfy one or more objectives: - To reduce the number of refund claims; - To speed up refund processing; - To address trader cash flow concerns, and reduce other costs of compliance; - To shield the refund system from abuse, and enhance the VAT revenue base generally; and - To reduce costs of tax administration. Observations from the source text: - Most schemes add to both administration costs and taxpayer compliance costs. - Only scheme D (preferential treatment of good compliers) achieves a reduction in these costs, while at the same time accelerating refund payments for most traders and enhancing protection of the VAT revenue base. - Scheme D tends to favor older established firms over new enterprises (but effects can be cushioned by education and assistance programs for new enterprises). - Two schemes (VAT bank accounts and large-scale invoice matching) have not proven benefits that justify their considerable additional costs. - Scheme A (zero-rating supplies to exporters) raises questions about whether real net benefits are delivered, given the complexity it adds to the VAT system. - Scheme C (certification of refund claims by CPAs) may be considered if traders accept additional costs and the accounting profession can be trusted, but should not substitute for development of tax administration capacity. - Schemes E (purchases paid through the banking system) and G (deferment of VAT on imported capital goods) may have a place in developing and transition economies, but should not be seen as longer-term solutions. ### Table 6 — Summary Findings (as presented) Performance criteria columns: Reduces or eliminates refund delay | Reduces number of refund claims | Enhances protection of VAT revenue base | Reduces taxpayer compliance costs | Saves admin resources - A. Zero-rated supplies to exporters: Yes | Yes | No | No | No - B. Large-scale cross-checking of invoices: No | No | Yes | No | No - C. Certification of refund claims by CPAs: Yes | No | Yes | No | Yes - D. Preferential treatment of good compliers: Yes | No | Yes | Yes | Yes - E. Purchases paid through banking system: Yes | No | Yes | No | No - F. VAT bank accounts: No | No | Not proven | No | No - G. Deferment of VAT on capital goods: Yes | Yes | No | No | No ### Suggested Model of Best Practice — Key Features and Recommendations - Maintain the number of VAT payers at a level that can be realistically managed by the tax administration. A high VAT registration threshold should be maintained until the tax authority is sufficiently developed to administer a larger number of VAT payers and refund claimants in a self-assessment environment. - VAT registration applications should be subject to proof of identity and other basic checks to prevent fictitious traders from entering the VAT system and stealing from the government through the refund system. - Establish suitable forecasting and monitoring systems to anticipate refund levels and make sufficient funds available to meet all legitimate refund claims when they occur. Authorities should be able to predict, with some degree of certainty, the level of refunds they might reasonably expect to pay throughout the year. - Refunds should be processed (i.e., paid, offset, or denied) within a reasonable statutory period (e.g., 30 days of the date on which a refund claim is made). The statutory deadline may be extended where: - (1) a filed VAT return is incomplete; - (2) the taxpayer has outstanding tax returns; - (3) the taxpayer has failed to respond within a reasonable period to verification enquiries; or - (4) the tax authority suspects, on reasonable grounds, that the VAT return is inaccurate and/or the taxpayer is engaged in fraudulent activity, in which case the taxpayer should be subjected to audit and/or investigations. - The tax administration should report publicly on its performance in meeting the statutory deadline for processing refunds. - Interest should be paid on late refunds to compensate taxpayers with legitimate refund claims for being deprived of their working capital. - Excess VAT credits should be offset against VAT and other tax arrears, except where an outstanding amount is subject to a genuine dispute. Necessary taxpayer accounting and debt management systems need to be in place. - Immediate refunds of excess VAT credits should always be paid promptly to exporters or to enterprises that export a large share of their products (e.g., where at least 50 percent of the turnover is attributed to export sales). As appropriate, other taxpayers may be required to carry-forward their excess credits for six months. If at the end of this period an amount of excess credit remains, that amount should be refunded to the taxpayer. - Verification of VAT refund claims should be part of a wider audit program aimed at broad coverage of taxpayers and compliance issues. Pre-refund audits should be limited to high-risk cases only (e.g., the first refund claim by a new registrant), while lower-risk claims should be subjected to selective post-refund audits. - Preferential treatment should be given to regular exporters with sound compliance histories. Examples include assigning an approved refund level within computer systems for taxpayers with sound compliance records and accounting practices, or categorizing refund claimants according to compliance history and perceived risk so that low risk claimants receive automatic refunds and higher risk taxpayers are required to substantiate claims. - Apply appropriate sanctions consistently to taxpayers who falsely claim refunds or do not comply with record-keeping requirements. Refund-related fraud should be prosecuted through the criminal justice system. - Ensure taxpayers can appeal, on reasonable grounds, decisions to withhold refunds. Appeals should be considered by an independent tribunal and dealt with promptly. - Provide clear information to taxpayers explaining their rights and obligations, and the procedures for making a refund claim. VAT returns and refund claim forms should be simple, have clear instructions, and be filed through means convenient to taxpayers. ### Conclusions (excerpted) - There is a tendency for tax administrations to deviate from the primary goal of building sound VAT systems based on improved voluntary compliance through effective systems of self-assessment when addressing VAT refund processing problems. - In reacting to attacks on the refund system, administrations can lose sight of longer-term objectives and introduce measures incompatible with self-assessment concepts, including intrusive and costly measures that apply to all taxpayers irrespective of compliance behavior. - Experience strongly suggests that the VAT is feasible only as a self-assessed tax, meaning that substitutes for effective risk-based approaches within a self-assessment environment cannot be expected to provide sustainable solutions to compliance problems related to VAT refunds. *Source: Box 5 and accompanying sections from the provided IMF document content.* --- _Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2005/_wp05218.pdf_