Annex I. Cross-Country Comparison of Special (Offshore) VDPs1 2

article image
article image
article image
article image
article image
Sources: Agenda Tributaria 2012; Martin and Camarda 2017; OECD 2010, 2015; and Transparency International 2017

Annex II. The Greek Voluntary Disclosure Program (VDP)

A. Key Design Features


  • After a (rather unsuccessful) tax amnesty asset repatriation scheme in 2010, a VDP was introduced in December 2016 for undeclared income and assets of previous years. The VDP initially applied until end of May 2017 and was subsequently extended until the end of November 2017.

  • The VDP did not amount to a tax amnesty and instead required the payment of taxes for undeclared income or assets based on the tax rates applicable on the year of the original tax liability. Repatriation of assets was optional.

  • The VDP was basically advertised as a favorable regime for late filing of initial or amending tax returns. There was no special application for entrance into the program; standard tax returns were filed.


  • Taxpayers (both individuals and legal entities) who had not filed tax returns or filed inaccurate tax returns for all taxes, duties, or levies and for whom (1) no audit order had been notified, (2) an audit order had already been notified, and (3) a temporary corrective tax assessment (after initial audit findings) had been notified.

  • Non-Greek entities with Greek beneficial owners were not eligible.

  • Submission of tax returns was possible regardless of the statute of limitations period.


  • Cases in which a final corrective (after-audit) tax assessment had been issued until December 12, 2016 (entry in force of the VDP law).

  • Ineligible tax returns: (1) returns filed under reservations, (2) returns that would result in refunds, and (3) income tax returns declaring losses in the respective tax year.

  • Amounts deriving from criminal activities (explicit reference to AML legislation), except for tax evasion as an underlying AML offense.

  • Politically exposed persons at any time in the past and their close relatives.

Incentives Provided

  • Significantly reduced fines (which were 60 percent-120 percent of the main tax for tax years until 2014 and 50 percent for tax years from 2014 onward): 8 percent through 36 percent of main tax, depending on how soon the taxpayer entered the program and whether an audit order had been issued. Interest was paid in full, where applicable (not applicable to tax liabilities until 2014).

  • No other administrative and/or criminal penalties were triggered for tax evasion, including prosecution for tax evasion as a predicating ML offense.

  • Any enforcement measures were withdrawn.

  • Entrance into the VDPs did not constitute an audit risk-management criterion.


  • Standard tax returns were filed in the local tax offices; no special VDP administration was created.

  • Supporting documentation was required, in line with requirements applicable at the time of the initial filing obligation. In the absence of supporting documentation, any other available evidence or supporting material could be submitted.

  • Possibility to file information/reporting returns (for example, real estate ownership returns that are used as a basis for the calculation of property taxes).

  • The total tax liability was payable within 30 days from the submission of the tax return.

  • Very limited possibilities to correct a tax return filed under the program—no possibility of refunds of set-offs against other tax liabilities.

  • The taxpayer could benefit from available debt settlement programs (for up to 24 installments).

B. Aspects of the Greek VDP Legislation That Were Difficult to Implement

  • The system was very complex and hence very difficult to administer. The calculation of the tax liability was burdensome, as there were various rates for various types of taxes to be applied. Endless calculation tables were devised, and consecutive circulars were issued, with several illustrative examples, to make the system administrable for the tax officers who were receiving the tax returns and calculating the tax liability.

  • Identifying the year when the tax liability was generated and determining the exact nature of the income, to apply the correct tax rates for the corresponding year, was challenging. Tax returns had to be accompanied by supporting documentation in line with the requirements applicable at the time of the initial filing obligation. This requirement created additional difficulty, as taxpayers could not provide the necessary documentation in all cases. The problem was solved by eventually allowing the submission of any other relevant evidence or material, in the absence of the required supporting documentation.

  • E-filing and e-assessment were not possible due to the complexity of the system as well as the limited IT capacity of the Greek tax administration. All tax returns had to be filed in paper form and the assessment had to be completed manually. This effort led to great delays in tax assessment; it was finally resolved by forming VDP assessment teams within the local tax offices.

  • Initially, there was no possibility to correct a tax return filed under the program. That situation created all sorts of problems with real estate ownership information returns, which sometimes had to be amended for corrections in the property description to match the property transfer deed. Because this effort led to significant blockages in real estate property transfers, the law was subsequently amended to allow for correcting tax returns in these cases.

C. Lessons from the Greek VDP

  • A good communication strategy is key. The program attracted much attention in the country It was widely advertised as the last opportunity for tax evaders to come clean in a far-ranging publicity campaign, which was developed as a joint initiative of the minister of finance, the tax administration, and the anti-corruption authority

  • The threat must be credible. The combination of the VDP with a highly visible crackdown on tax evasion made the program an effective instrument. The Greek tax administration’s information powers as well as audit capabilities (new risk-management system) had already been enhanced after the introduction of the new tax procedures code in 2014. In parallel, the tax administration was prominently advertising a newly developed system for the cross-matching of bank accounts with tax returns of approximately 1,270,000 taxpayers located throughout the so-called evasion lists (lists of individuals’ foreign bank accounts for the years 2002–14) obtained through various foreign sources. Also, Greece was in the process of implementing the EU and international framework on exchange of information, with various legislative acts adopted within the first months of 2017. Finally, taxpayers under audit received letters notifying them about deadlines for entrance to the program, which confirmed that the audit was “after them.”

  • Simplicity is efficiency. The complex system of calculations could not be supported by the development of an IT system in the tight deadlines provided. The tax assessment was performed manually, thus posing a huge administrative burden for the already stretched tax administration.

  • Collection should be swift. The possibility for installments posed high risks for collection, and some taxpayers were excluded from the program for not following up on their payment obligations—the law provided for the reversal of the offered incentives in this case.

Annex III. Administrative Lessons from South African Offshore Voluntary Disclosure Program (VDP)

Some of the aspects of the South African Offshore VDP that were found to be impractical and impeded the effective functioning of the VDP unit were the following:

  • The initial application period was too tight, as many applicants had to obtain the required documents themselves. This period was subsequently extended, which was beneficial in the following ways:

    • Some approvals of VDP were granted, and applicants received amnesty as promised without any negative consequences for them. Word spread, enticing more people to apply. So, if applications remain open once initial approvals are granted, more VDP applications will come. News that the VDP administration supporting tax measures was reasonable helped spread the word that the process is fair, and prospective applicants developed the necessary trust in the system.

    • More than half of all the 43,000 applications were received in the last week of the VDP period, including the final day (Sunday) until midnight.

  • Thus, it is important to allow the assigned VDP administration through its actions to earn the confidence of the public.

  • The legislation as promulgated made no provision for individual staff members to consider applications; initially, each application therefore had to be considered in a full unit meeting with all unit members attending. This process was untenable and very slow. The matter was subsequently amended, with each application thereafter considered by a two-person team comprising a representative of the South Africa Reserve Bank (SARB) and the SA Revenue Service.

  • The stipulated payment period of three to six months was problematic. Three months were simply too short and even six months became challenging. A period should be provided in legislation, but the unit should have the discretionary power to extend it if informed by objective criteria or facts. For example, banks make serious mistakes—to the detriment of VDP applications. According to legislation, these mistakes could have meant that applicants lost their “amnesty” status, owing to nonpayment or because of incorrect payment by banks. So, it is important that all applications must be processed within the time set aside for the VDP; however, an extended VDP period of 10 to 12 months may be necessary. Due diligence and verification of asset registers are time-consuming efforts, especially since they involve data sharing and coordination with competent authorities in other tax jurisdictions.

  • Refund provisions for paid fees and applications were problematic. Initially no one expected refund requests, but these related to observed errors by financial intermediaries (for example, banks).

  • Applicants’ refund requests raised suspicion among VDP unit members, who considered a refund from the SA Reserve Bank to be vulnerable to money laundering risks. However, applications could show, for instance, that banks paid the same amount twice to the adjudication unit. Experience indicated that refund requests were never for large amounts; they merely pointed to banks’ clerical errors.

  • Initially, only individuals, not companies or trusts, could apply for the VDP. This rule was too limiting, especially where personal affairs of applications were intertwined with private companies or trusts. Planning for a more general VDP would have taken longer but would have worked better.

  • It was necessary to issue provisional and final letters to applicants, with provisional letters stating the VDP application is to be finalized once the VDP levy payment is received. These letters stated that “amnesty” was granted only based on the disclosure made by applicants in respect of the original sources of funds that led to the creation of foreign assets; however, this stipulation resulted in many complex administrative requirements.

  • In 2003, digital capacity was underdeveloped and at the time, VDP applications were paper based. In practice, 43,000 applications had to be filed and locked away. Hence, any VDP approval should be electronic, rather than paper based, with a system that generates electronic response letters and other correspondence.

  • Taxpayers who were already under investigation by the exchange control department of the SARB could not qualify for the VDP. This situation became an administrative hurdle, as exchange control records were inadequate, with missing names of listed investigation targets when reported to the adjudication unit. A further complication was that the SARB exchange control department could continue issuing letters of investigation without verifying that investigation targets may already have filed a VDP application. So, some taxpayers submitted VDP applications and then received letters of investigation, unrelated to their applications, which led to some suspicion and threatened the uptake ratio of the VDP. With hindsight, the administrators felt that the VDP should therefore cover people under investigation or potentially under investigation.

  • Owing to the volume of applications, the unit existed much longer than initially envisaged—and hence, the advice is to provide for a say 10-month period for submission of VDP applications but extend the duration of the VDP unit’s review, adjudication of applications, and audit of VDP flat tax payments for more than a year.

Note that the South Africa VDP of 2003 predates FATF’s policy; hence, the policy was never reviewed hy FATF or hy the Eastern and Southern Africa Anti-Money Laundering Group, the relevant regional body of which South Africa is also a member. Such a review by the FATF or the regional body is a current requirement of FATF, to ensure that countries comply with the international AML/CFT standards. Therefore, countries that copy the 2003 South Africa VDP model may not receive a stamp of approval by FATF.


  • Agencia Tributaria. 2012. Tax Revenue Annual Report—2012.

    • Search Google Scholar
    • Export Citation
  • Alstadsæter, A., N. Johannesen, and G. Zucman. 2019. “Tax Evasion and Inequality.” American Economic Review 109 (6): 2073103.

  • Baer, K., and E. Le Borgne. 2008. Tax Amnesties: Theory, Trends, and Some Alternatives. Washington, DC: International Monetary Fund.

  • Beer, S., M. Coelho, and S. Leduc. 2019. “Hidden Treasure: The Impact of Automatic Exchange of Information on Cross-Border Tax Evasion.” IMF Working Paper 19/286, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Bethmann, D., and M. Kvasnicka. 2017. “International Tax Evasion, State Purchases of Confidential Bank Data and Voluntary Disclosures,” Beiträge zur Jahrestagung des Vereins für Sozialpolitik 2017: Alternative Geld- und Finanzarchitekturen – Session: Taxation II, No. B19-V3, ZBW -Deutsche Zentralbibliothek für Wirtschafts-wissenschaften, Leibniz-Informationszentrum Wirtschaft, (Kiel, Hamburg).

    • Search Google Scholar
    • Export Citation
  • Financial Action Task Force (FATF). 2012Managing the Anti-Money Laundering and Counter-Terrorist Financing Policy Implications of Voluntary Tax Compliance Programmes.” In Annual Report 2012–2013. Paris: Organisation for Economic Co-operation and Development.

    • Search Google Scholar
    • Export Citation
  • KPMG. 2015. Tax Amnesty – A Viable Option for Boosting Revenue.

    • Search Google Scholar
    • Export Citation
  • Mastellone, P. 2015. “The Italian Voluntary Disclosure Programme: A New Era of Tax Amnesty?European Taxation 55 (8): Journal Articles and Papers IBFD.

    • Search Google Scholar
    • Export Citation
  • Martin, L., and A. Camarda. 2017. “Best Practices in Tax Amnesty and Asset Repatriation Programmes,” Transparency International.

  • Mattiello, G. 2005. “Multiple Tax Amnesties and Tax Compliance (Forgiving Seventy Times Seven).” Working Paper 06/2005, Universita Ca’ Foscari, Venice.

    • Search Google Scholar
    • Export Citation
  • Mikesell, J. L., and J. M. Ross. 2012. “Fast Money? The Contribution of State Tax Amnesties to Public Revenue Systems.” National Tax Journal, Sep 2012 65 (3): University of Chicago Press (Chicago).

    • Search Google Scholar
    • Export Citation
  • Organisation for Economic Co-operation and Development (OECD). 2007. Improving Access to Bank Information for Tax Purposes.

    • Search Google Scholar
    • Export Citation
  • Organisation for Economic Co-operation and Development (OECD). 2010. Offshore Voluntary Disclosure—Comparative Analysis, Guidance and Policy Advice. Paris.

    • Search Google Scholar
    • Export Citation
  • Organisation for Economic Co-operation and Development (OECD). 2015. Update on Voluntary Disclosure Programmes: A Pathway to Tax Compliance. Paris.

    • Search Google Scholar
    • Export Citation
  • Rossouw, J. 2006. “The Exchange Control and Related Tax Amnesty Process,” Bankindaba, the staff magazine of the South African Reserve Bank (Pretoria).

    • Search Google Scholar
    • Export Citation
  • TAS Research and Related Studies. 2017. An Analysis of Tax Settlement Programs as Amnesties, Vol. 2.

    • Search Google Scholar
    • Export Citation
  • Waerzeggers, Christophe, Cory Hillier, and Irving Aw. 2019, “Designing Interest and Tax Penalty Regimes.” Tax Law IMF Technical Note 1/2019, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation

EOI effectiveness outside the automatic EOI (AEOI) framework is doubtful because, under the EOI on request framework, provided for by tax treaties, information can be requested only about a specifically named person and to the extent that the information is “foreseeably relevant” to the administration, enforcement or recovery of any taxes, a so-called “fishing expedition” for taxpayer data is not possible. However, it should separately be noted that, in some countries, police can request from any bank any information also at the pre-investigation stage (for example, a fishing expedition), which is in line with Financial Action Task Force [FATF] Recommendation 31 (that is, Assessment Criterion 31.3.a), and includes countries that have centralized bank account registries that law enforcement and other competent authorities can consult to request information. And most often and in addition, law enforcement will be able to use the financial intelligence unit to ask banks for specific data. Finally, in some Nordic countries, individual tax information is published, and foreign authorities can combine these with domestic data. Beer, Coelho, and Leduc (2019) also find no effect of EOI, but strong effect of AEOI.


The US Foreign Account Tax Compliance Act generally requires foreign financial institutions and certain other nonfinancial foreign entities to report on the foreign assets held by their US account holders.


Not all VDPs insist on asset repatriation; some allow the maintenance of any foreign asset holdings, but with due payment of all arising tax liabilities.


According to the Organisation for Economic Co-operation and Development (OECD 2015, para. 18), “Typically, countries do not waive tax as part of their voluntary disclosure programme. Waiving tax would represent some form of a tax amnesty.” The OECD report indicates that most countries that have implemented a VDP require the payment of tax upon making a qualifying voluntary disclosure.


For example, the Common Reporting Standard Multilateral Competent Authority Agreement, or the EU Directive on Administrative Cooperation (2011/16/EU, as subsequently amended).


Martin and Camarda (2017) and OECD (2015, 29–129) reviewed VDP legislation for these countries: Argentina, Australia, Austria, Belgium, Canada, Chile, People’s Republic of China, Costa Rica, Croatia, Czech Republic, Denmark, Estonia, Finland, France, Germany, Hungary, Iceland, India, Indonesia, Ireland, Italy, Japan, Jersey, Korea, Latvia, Liechtenstein, Lithuania, Luxembourg, Malaysia, Malta, Mexico, Netherlands, New Zealand, Norway, Poland, Portugal, Russian Federation, Singapore, Slovak Republic, Slovenia, South Africa, Spain, Sweden, Switzerland, Turkey, United Kingdom, and United States.


Several countries, such as Albania, Greece, Italy, Portugal, Slovenia, and Spain, have enacted VDPs in which legal entities were entitled to benefit from the program.


The VDP cannot grant the taxpayer immunity from investigation, prosecution, or conviction for money laundering or a predicate offense in relation to declared or repatriated funds or other assets, nor can the assets be immune from confiscation should they be proved to be proceeds of crime.


The Dutch VDP successfully applied this principle—that is, those who are already under investigation can be excluded from the VDP. The principle holds that, “Voluntary disclosure must take place before we know what has not been disclosed.” The amount of the fine also depends on how long ago you failed to file a return or filed an incorrect one. See


It can also create practical difficulties, as the experience in South Africa showed, where proof of the taxpayer’s knowledge of being under investigation at the time of applying for the VDP became a procedural sticking point. A condition of this nature, therefore, is feasible only to the extent that the VDP unit can verify unequivocally that the taxpayer was in fact under investigation upon receipt of the VDP application. Alternatively, penalty waivers for taxpayers/prospective VDP applicants could be structured in such a way that they are subject to a significantly higher penalty if already under audit at the time of application.


Taxpayers should not be exempt, by law or in practice, from investigation, prosecution, or conviction for money laundering/financing of terrorism (ML/FT) in relation to repatriated funds or other assets, or for confiscation of these assets if these are proceeds of crime (even if the taxpayer had paid taxes over these illicit income or assets).


There is a legal difference between a tax authority and/or the prosecution to have a policy in place not to prosecute a certain conduct if certain conditions are met, and on the other hand, “immunity” which is more intrusive as it is a prohibition in law to prosecute someone, regardless of other factors (for example, a head of state may be immune). Hence, a VDP should not grant an immunity from prosecution for money laundering and any predicate offenses because immunity often extends to investigations (in law or in practice) and prevents in law or practice legal actions against enablers that may legally not be covered by the immunity.


The FATF defines PEPs as individuals (including family members or close associates) who are or have been entrusted with prominent public functions by a country (for example, heads of state or of government, senior politicians, senior government, judicial or military officials, senior executives of state-owned corporations, important political party officials), and distinguishes foreign from domestic PEPs. It also recognizes a third category of PEPs—namely, persons who are or have been entrusted with a prominent function by an international organization (for example, members of senior management, directors, deputy directors, and members of the board or equivalent functions). The definition of PEPs is not intended to cover middle-ranking or more junior individuals in the foregoing categories.


It should be noted that the information in Annex I relates to case studies and is not included as models of best practice.


Not all countries with fundamental AML/CFT shortcomings will be subject to gray listing, especially not countries with smaller economies.


See FATF Best Practices, Principle 1: Effective application of AML/CFT preventive measures.


FATF Best Practices: Principles 3 and 4.


According to the 2015 OECD Update on VDPs (p. 8), “Taxpayers are also concerned about the confidentiality of the information that is provided, both because of the reputational damage that might result from any publicity and for reasons of personal security. Taxpayers want reassurance that the financial terms on which their liabilities will be settled will not be prohibitive. They also want reassurance that once the disclosure is complete, they will not be unduly targeted for enhanced scrutiny in the future.” Issues of confidentiality and use of information disclosed should therefore be based on legal obligation and the laws of a country and should not be guided by discretions or verbal commitments by officials. In countries where adherence to confidentiality standards is a concern, practical solutions were adopted in some VDPs such as strictly limiting access to provided information in VDP applications to designated tax officials who can protect the information via special legislative tax secrecy provisions applying to tax officials in the VDP unit. Of course, transparency as to the use of provided VDP information should still be guaranteed and VDP information must be shared among relevant organs of state (that is, the FIU).


FATF Recommendations 9, 10, and 20.


Taxes, including a VDP flat tax, penalties, and interest.


Some VDPs specify that any party or resources that assisted with the assets’ acquisition need not be revealed.


Indicators in this regard can be FATF or FSRB assessments and follow-up reports, and relevant ratings from these reports. For example, a country that is subject to FATFs enhanced monitoring, or is in FATF or FSRB enhanced monitoring, may have difficulties establishing the ability to mitigate ML/FT-related VDP risks despite established ML/FT shortcomings. But even countries that have generally good ratings may have crucial weaknesses in their system—for example, if domestic cooperation and information exchange is not sufficient, or if the ML criminalization has shortcomings, including if self-laundering is not criminalized.


Note that, from a strictly legal perspective, it would only be a criminal court that can establish crime; and any suggestion that this function of a criminal court could be delegated to a special unit is not supported by the IMF’s Legal Department since there is the need for a legal basis to do this, even under administrative law.


Annex I preserves the original naming of the program—that is, a program is called “tax amnesty” even if it essentially constitutes a VDP.


Reporting on these country examples should not be construed that the IMF approves of the adopted tax amnesties or VDPs—much depends on their design and the adherence to AML/CFT provisions.


When available.

Voluntary Disclosure Programs — Design, Principles, and Implementation Considerations
Author: Ms. Dora Benedek, Martin Grote, Grace Jackson, Maksym Markevych, Mr. Christophe J Waerzeggers, and Ms. Lydia E Sofrona