Chapter 5. International aspects of the taxation of household savings

This chapter discusses the taxation of household savings from an international perspective, and considers the impact of exchange of financial account information between tax administrations on the taxation of capital income internationally. The chapter discusses the recent adoption and expansion of exchange of information in detail, highlighting the dramatic increase in the coverage of exchange of information networks. The automatic exchange of information is likely to make it harder for taxpayers to engage in tax evasion and may reduce the distortions involved in levying taxes on capital income. The chapter also highlights potential challenges that remain with respect to ensuring the coherent taxation of capital income on a residence basis in an international context.


5.1. Introduction

This chapter discusses the taxation of household savings from an international perspective, and considers the impact of exchange of information (EOI) on the taxation of capital income internationally. Previous chapters considered effective tax rates across asset types from the perspective of taxing a domestic saver. However, in an increasingly globalised economy a large share of total investment comes from abroad and, equally, a large share of domestic savings is invested abroad. This means that the returns to investment domestically often accrue to foreign investors. It also means that the capital income earned by domestic residents is often earned abroad.

This chapter focuses on the taxation of foreign-sourced capital income at the individual level levied in the taxpayer’s tax residence. A crucial component of effectively implementing residence-based taxation of foreign income is the effective EOI between tax authorities. Without effective EOI, it is possible that some taxpayers may fail to declare their foreign-sourced income or assets and, through a lack of information in the hands of the tax authorities, avoid detection. This policy challenge has received widespread attention over recent years.

Effectively implementing international EOI has long been on the agenda of tax policymakers. However, some early efforts to exchange information have faced a number of challenges. In some instances, limited membership of countries in an EOI network has led to the shifting of assets to other countries not in the network (i.e. countries who have not signed an EOI agreement). In other instances, certain income or asset streams have been left out of information exchange agreements, leaving open some channels of evasion for taxpayers. Finally, information exchange on request has required tax authorities to already have some information prior to seeking information through formal information exchange channels from their foreign counterparts.

This chapter discusses the recent expansion and standardisation of EOI in detail. The implementation of the Common Reporting Standard (CRS) has allowed for EOI on an automatic and standardised basis across a rapidly expanding number of countries. This chapter highlights the dramatic change in the coverage of EOI networks, and the promulgation and monitoring of this new standard to improve the quality of information exchanged. It places these new initiatives in the context of previous EOI efforts, and highlights the potential impact of EOI for the taxation of capital income at the individual level overall.

This chapter also highlights potential challenges that remain with respect to ensuring the coherent taxation of capital income on a residence basis in an international context. These challenges are broadly twofold. The first relates to ensuring that information exchange is effectively implemented. This chapter highlights the importance of ensuring a comprehensive EOI network amongst all relevant jurisdictions. This is necessary to prevent assets from flowing to jurisdictions that remain outside the global EOI network. It is also important that the CRS is implemented consistently across all jurisdictions that adhere to its principles. These tasks have been assigned to the Global Forum on Transparency and Exchange of Information for Tax Purposes or “Global Forum”1 and highlight the need not only for a robust Global Forum peer review process but also for technical support, particularly for countries and jurisdictions with limited administrative capacity. It is also important to ensure that persons, assets, and institutions not covered under existing EOI standards do not offer opportunities for continued tax evasion and therefore that the standards remain fit for purpose.

The second challenge relates to coherently taxing capital income in a world where EOI is effectively implemented. First, the impact of EOI will be limited unless tax authorities have the administrative means and methods to effectively use the information exchanged. Tax authorities should take advantage of new analytical tools and technological advances. Second, the expansion and effectiveness of EOI may also induce taxpayers to shift their wealth to assets not covered by EOI, such as real property. This increases the importance of the tax treatment of these assets. Third, taxation at the personal level in a world of falling corporate tax rates may lead taxpayers to shift income from the personal level to the corporate level, and may also exacerbate lock-in problems as taxpayers postpone the realisation of capital gains. This suggests that a key policy priority is ensuring that the corporate income tax functions effectively as a “withholding” tax for what is in effect personal capital income. Taxpayers may also “vote with their feet” as a result of countries’ efforts to strengthen the taxation of capital income in their tax residence, although these mobility responses, which are worthy of more detailed analysis, are not considered in this chapter.

The chapter proceeds as follows. Section 5.2 discusses key choices available to international tax policymakers and considers the development of the international tax system over recent years. Section 5.3 discusses recent developments in exchanging information between tax authorities in more detail, focusing on the development of the Exchange of Information on Request (EOIR) networks in recent years as well as the expansion of Automatic Exchange of Information (AEOI) through the development of the CRS. Section 5.4 considers the impacts of the expansion of EOI between tax authorities. Section 5.5 concludes with some potential priorities for policymakers following the advent of widespread EOI.

5.2. Capital taxation and globalisation

Globalisation presents a wide variety of policy challenges, not least in the area of savings taxation (OECD, 2017a). Many of the key tax policy insights that can be drawn from the optimal taxation literature about the design of capital income taxation in a social welfare maximising setting have been developed from the perspective of a closed economy.2 However, continuing expansion of global trade in goods and services as well as expansion of cross-border investment and financial flows mean that national tax policies cannot be designed in isolation from one another.

This globalisation of the world economy has been intensifying over recent decades. For the purposes of capital taxation, the key trend brought about by globalisation has been the increase in cross-border financial and investment flows. Foreign investment is an increasingly large share of total investment in any given economy and forms a larger share of the capital tax base in any given economy. Savings and ownership are now, more often than not, cross national; a taxpayer does not need to be tax resident in a country to invest there. In addition, a taxpayer may invest in the country where they are tax resident, but mediated through foreign companies or structures.

Strong tax policy rationales exist for the taxation of capital income at source. Source-based taxation implies that the returns to capital are taxed only in the country where value is created. Source-based taxation allows countries to tax location-specific economic rents and compensates the source jurisdiction for the public services that it provides to investors. Source-based taxation also contributes to the coherent functioning of the domestic tax system in that it provides for a withholding function of the taxes on capital income at the individual level.

However, as countries tax the income earned by the capital being used in their countries at source, the cost of capital will rise and domestic investment will fall. It may negatively impact the inflow of FDI and will stimulate the outflow of domestic savings. This in turn will reduce labour productivity and so will reduce wages. It is widely documented that source-based taxes are partially borne by immobile factors (such as land or labour) in that country. The academic literature estimates that between 30% and 55% of international corporate income taxation is borne by labour (Milanez, 2017; Fuest, Peichl and Siegloch, 2018; Arulampalam, Devereux and Maffini, 2012). This raises the question of how and on what basis internationally mobile capital is best taxed.

The challenges of taxing mobile capital in an internationally competitive tax environment are evident in the trends in both the average combined corporate income tax (CIT) rate, and the average combined statutory rate on dividends in recent years. The total tax burden on dividend income has increased slightly since the crisis, in contrast with the pre-crisis period during which the tax burden on dividend income generally decreased. The post-crisis rise in dividend taxation has been the result of increased taxation at the shareholder level while corporate tax rates continued to decrease modestly, as shown in Figure 5.1. The figure shows the continuous fall of the average standard CIT rate in OECD countries from 32.5% in 2000 to 24.7% in 2016. This trend towards lower CIT rates has continued in the aftermath of the OECD/G20 BEPS project, which has levelled the playing field by preventing harmful tax competition in relation to the tax base. These reductions in taxation at the corporate level highlight the importance of taxation through the PIT system, which is the focus of this study.

Figure 5.1. Trends in Corporate Income Tax Rates

Source: OECD Tax Database

Discussions about the optimal design of capital income taxes often focuses on efficiency issues, but equity considerations are equally important. From an equity perspective, it is the overall tax burden on capital income that matters, irrespective of whether capital income is taxed at source or residence. Residence-based taxation implies that the returns to capital taxes are levied where the investor is tax resident. These taxes are in theory levied whether the income is earned on capital invested at home or abroad. It ensures that tax residents who earn their income in the form of labour or capital income are taxed on a relatively equal footing, thereby strengthening the fairness of the tax system.

However, taxing foreign source income on a residence basis is complicated by the challenge of tax evasion. Individuals who do not report their foreign source income or assets to their domestic tax authority can evade paying taxes on this income if the domestic tax authority cannot access information about this income or these assets. Estimates of the size of offshore wealth are challenging to construct and maintain, and are subject to considerable uncertainty. However, recent estimates in the literature have varied from USD 6-7 trillion to USD 22 trillion (Alstadsaeter et al., 2017). The general order of magnitude of these estimates suggests that levying residence-based personal taxes is a considerable policy challenge.

Information exchange is a key aspect of the solution to international tax evasion, and, therefore, information exchange provisions have been a key aspect of tax treaties, as is discussed below. Effective information exchange involving foreign-source income is crucial for the maintenance of residence-based taxation and potentially for capital taxation in general. Without adequate information about cross-border income and wealth, it is all too easy for taxpayers to conceal wealth from the tax authorities of their residence jurisdiction and avoid detection. Moreover, as the combination of source and residence taxation may imply that capital income is unintendedly taxed twice, effective information exchange is also crucial in order to prevent double taxation.

5.3. Exchange of information


This section outlines recent progress made towards effective EOI between countries. Key initiatives discussed include the application of a common international standard on EOIR, and more recently AEOI. The development of these EOI mechanisms allows countries to better cooperate with one another to detect and identify capital assets and income that could otherwise be hidden.

The key forum for promoting the widespread implementation of EOI internationally is the Global Forum on Transparency and Exchange of Information (The Global Forum). The Global Forum was created in the early 2000s to address the risks to tax compliance posed by non-cooperative jurisdictions. On 2 April 2009, following a call from the G20 to “end bank secrecy”, the Global Forum moved to strengthen and expand EOI on a world-wide basis (G20, 2009). Today, with over 147 member countries and jurisdictions, it is the key global body ensuring international tax co-operation through the effective implementation of the international standards of tax transparency and exchange of information (i.e. EOIR and AEOI).3

EOIR and AEOI are complementary tools in the fight against tax evasion. Effective AEOI will provide tax administrations with information needed not only to tax the foreign source income and capital of their tax residents that is located abroad, but also with the necessary information to effectively use the OECD standard on EOIR. As discussed, tax administrations need specific information (e.g. information identifying the taxpayer and the jurisdiction where assets may be located) in order to file a request for information (see Section 3.3 above). With the information obtained from the AEOI, tax administrations will have details to file a request. As a result, both standards are interdependent, working together to enhance the effectiveness of tax administrations’ efforts in addressing international tax evasion.

Multilateral approaches to EOI

Two key instruments promulgated by the Global Forum to facilitate information exchange include the Convention on Mutual Administrative Assistance in Tax Matters (the Convention) and the OECD’s Model Agreement on Exchange of Information for Tax Matters. The Convention is the most comprehensive multilateral instrument available for AEOI.4 In 1988, the Council of Europe and the OECD jointly developed the Convention as a multilateral instrument among OECD member countries, allowing all forms of tax co-operation including AEOI. In 2010, the Convention was amended to allow it to become open to non-OECD countries. It allows for administrative assistance across all types of taxes, facilitates AEOI and EOIR, as well as spontaneous exchange, provides for the possibility of running cross-border and simultaneous tax audits, and allows states to assist each other in tax collection. The Convention also has the advantage that signature of the Convention creates EOI relationships with every other signatory, obviating the need for separate bilateral agreements in each instance.

The OECD’s Model Agreement on Exchange of Information for Tax Matters (OECD Model TIEA) may also be used for AEOI purposes. The TIEA was developed by the OECD in 2002 to promote international co-operation in tax matters. In its first version, it included the possibility of EOI considered “foreseeably relevant” to the determination, assessment, collection, recovery and enforcement, or the investigation or prosecution of tax matters under the EOIR. In 2015, the model TIEA was amended to include AEOI and spontaneous EOI. Even though a TIEA can either be bilateral or multilateral, only the bilateral version is currently being used in practice. In this instance, both countries exchanging information on an automatic basis must sign a TIEA containing provisions relating to AEOI with each other.

A key difference between the Convention and a bilateral Double Tax Convention (DTC) or TIEA is that countries signing the Convention provide the basis for AEOI with every other Convention signatory. By contrast, bilateral TIEAs require separate agreements with each country.

The scope of the Convention is broader than that of the model TIEA. One of the main differences between the model TIEA and the Convention is that the former looks to promote international co-operation in tax matters through EOI only, while the latter also includes other types of international co-operation such as assistance in recovery and simultaneous tax audits. Similarly, the Convention provides for assistance with respect to taxes like social security contributions (SSCs) which are not covered by the model TIEA. However, AEOI can take place under either the Convention or a bilateral TIEA following the OECD Model TIEA.

Exchange of Information on Request


EOIR refers to a situation where the tax authority of one jurisdiction asks for particular information from the authority of another jurisdiction (with which it has signed a tax treaty and/or a tax information exchange agreement) in connection with a tax inquiry or investigation (OECD, 2016). The requesting jurisdiction must provide specific details on the information it needs, and show that the requested information is “foreseeably relevant” to the tax affairs of a given taxpayer or a group of taxpayers in order for the other tax administration to answer the request.

The Global Forum’s EOIR standard is the most widely-implemented standard on EOI for tax purposes. The EOIR standard is primarily drawn from the 2002 OECD Model Agreement on Exchange of Information on Tax Matters (OECD Model TIEA) and its commentary, Article 26 of the OECD Model Tax Convention on Income and on Capital and its commentary, as updated in 2012 (OECD, 2016), and Article 26 of the UN Model Double Tax Convention.

EOIR is possible because there are domestic laws enabling availability of information, and there are international agreements allowing the exchange of that information between countries and jurisdictions.5 Jurisdictions implementing the standard are expected to introduce all necessary domestic regulation ensuring availability of the requested information as well as access to that information by tax/competent authorities of the requested jurisdiction. They are also expected to have international instruments in place that enable them to exchange information with all relevant partners.

Different mechanisms exist enabling EOI between jurisdictions. Some agreements are bilateral (i.e. an agreement taking place between two jurisdictions), while there is also an increasing use of multilateral agreements (i.e. agreements under which jurisdictions agree with many countries to exchange information). The various types of agreements that can allow for EOI are outlined in Table 5.1. Most double tax treaties include provisions which require EOI. For example, article 26 of the OECD Model Tax Convention provides the most widely accepted legal basis for bilateral EOI for tax purposes (OECD, 2016). Similar provisions can be found in other instruments such as the Convention and the OECD Model TIEA.

Table 5.1. Kinds of agreements providing for EOIR

Bilateral Instruments

OECD Tax Model (article 26)

OECD’s model agreement on EOI for tax matters (TIEA)

Tax treaties not based on the OECD model

Multilateral Instruments

The Convention on Mutual Administrative Assistance in Tax Matters (The Convention)

Other Multilateral Agreements (e.g. Multilateral TIEA)

EU Council Directive 2011/16

The development of the network of EOIR agreements

There has been a substantial increase in the number of agreements facilitating EOIR worldwide in the period since the 2009 G20 declaration. Figure 5.2 shows the development of the network of relationships for EOIR over this period. The panel (a) shows the network of information exchange relationships in 2009 amongst OECD and G20 countries. Blue areas have active EOIR relationships, grey areas do not. It is clear from the first figure that many gaps existed in the network of EOIR relationships amongst OECD and G20 countries in 2009.

Figure 5.2. The EOIR Network among OECD/G20 Countries, 2009 and 2018

Note: This graph includes EOIR relationships that may not meet Global Forum standards. Agreements are shown based on the earliest date of signature, ratification, or entry into force that is available from the Global Forum.

Panel (b) of Figure 5.2 shows the position of information exchange relationships in 2018. Here it can be seen that all OECD/G20 countries now participate in EOIR with each other. In large part, this change has come about not through increased numbers of bilateral relationships, but rather due to an increase in the number of countries that have signed the Convention.

Figure 5.3 shows a similar picture to Figure 5.2, but shows the network of EOIR relationships between OECD/G20 countries and a set of International Financial Centres (IFCs).6 It can immediately be seen from panel (a) that the network of EOIR relationships was far less comprehensive than was the case amongst the OECD/G20 countries in 2009.

Figure 5.3. The EOIR Network between OECD/G20 and OFCs, 2009 and 2018

Note: This graph includes EOIR relationships that may not meet Global Forum standards. Agreements are shown based on the earliest date of signature, ratification, or entry into force that is available from the Global Forum. Data on International Financial Centres are based on IMF (2004).

EOIR peer review process

The Global Forum seeks to ensure that its members fully implement the international standards of transparency on EOIR and AEOI through in-depth peer review and monitoring processes. The peer reviews of EOIR, which have taken place since 2009, involve a two-phase process under which an assessment team made up of peers and a member of the Global Forum Secretariat analyses the compliance of jurisdictions with the international standard of EOIR. The reviews are then adopted by the Global Forum’s Peer Review group before being adopted by all members of the Global Forum operating on an equal footing. Phase one reviews cover the legal and regulatory framework, while phase two looks into the implementation of the framework in practice. At the end of the process, each jurisdiction receives an overall rating of compliance, consisting of either: compliant, largely compliant, partially compliant, or non-compliant. A first cycle of EOIR reviews took place from 2010 until 2016.

The second round of EOIR reviews will assess both the legal and regulatory framework as well as the working of the standard in practice (i.e. the two phases of the reviews have been combined). This round has already been launched with the first ten reports being published in August 2017. This round is scheduled to run from 2016 until 2020 and reviews will be carried out against an enhanced EOIR standard which now includes requirements relating to the access to and exchange of beneficial ownership information (OECD, 2016). The ultimate goal of these reviews is to ensure jurisdictions effectively implement the international EOIR standard (OECD, 2016).

The peer-review process focuses on three main aspects: first, availability of information; second, access to information; and third, exchange of information. For EOIR to be effective, each jurisdiction must make sure that information (ownership information of legal entities and arrangements, accounting information and banking information) should be available and accessible to its competent authority.7 Finally, each jurisdiction should have appropriate international EOI instruments in place with all relevant partners. Information that is not available or cannot be accessed cannot be exchanged. Even if a jurisdiction never exchanges information, implementing the Global Forum’s standards on availability of and access to information is vital to ensuring that it can protect its own domestic tax base.8

The standard on EOIR provides for the exchange of foreseeably relevant information and does not allow for the possibility of “fishing expeditions”. This means that a jurisdiction looking for information must know which jurisdiction to ask for the information, provide sufficient information to identify the tax payer(s) and establish foreseeable relevance to a particular tax investigation. Furthermore, knowledge of the financial institution in a given jurisdiction that holds the information is typically required.

Automatic Exchange of Information


The OECD and G20 have also more recently developed an international standard on AEOI relating to financial accounts (consisting largely of the CRS which sets out the information to be reported by financial institutions and the due diligence obligations to ensure those financial institutions properly identify RFAs). This allows for international tax co-operation beyond what was possible under the EOIR. Based on an automatic and standardised methodology explained further below, AEOI provides for jurisdictions to periodically and automatically exchange information and obtain data about tax residents with financial assets abroad. Instead of jurisdictions requesting information on specific details about a specific taxpayer or establishing foreseeable relevance to a specific ongoing case, the scope and content of the information exchange is agreed in advance and the information is automatically exchanged each year. By the end of 2017, 49 member jurisdictions should have commenced exchanging information automatically under the CRS. A further 53 jurisdictions are due to commence in 2018.

A variety of mechanisms exist enabling AEOI between jurisdictions. As with EOIR, these can be either bilateral (e.g. a bilateral TIEA or a DTC) or multilateral (e.g. the Convention) (see Figure 5.4).9 The kinds of international agreements that can allow for AEOI are outlined in Table 5.2. Most AEOI takes place under the Convention, but other forms also exist.

Figure 5.4. International Instruments for EOI

Note: This chart only refers to forms of EOI that pertain to personal taxation. Other forms of information exchange such as country by country reporting may also exist. Abbreviations are as follows: FATCA IGA: Foreign Account Tax Compliance Act Intergovernmental Agreement; TIEA: OECD Model Agreement on Exchange of Information on Tax Matters; The Convention: The Convention on Mutual Administrative Assistance in Tax Matters; MCAA: The Model Competent Authority Agreement that specifies that information will be exchanged under the CRS which can be bilateral (BCAA) or multilateral (MCAA); CRS: The Common Reporting Standard.

Table 5.2. Primary Authoritative Sources – Agreements providing for AEOI

Bilateral Instruments

Foreign account tax compliance act intergovernmental agreement (FATCA IGA)

OECD Tax Model (article 26)

OECD’s model agreement on EOI for tax matters (TIEA)

Tax treaties not based on the OECD model (e.g. UN Tax Model Article 26)

Multilateral Instruments

Multilateral Instruments

The Convention on Mutual Administrative Assistance in Tax Matters (The Convention)

Other Multilateral Agreements (e.g. Multilateral TIEA)

EU Directive 2011/2016

Other forms of AEOI

Under the auspices of the European Union, AEOI has also been taking place since the early 2000s. In 2003, under the EU Savings Directive, Member States agreed to exchange information on interest payments derived from savings schemes (EU Council Directive 2003/48/EU).10 In 2011, under the Directive on Enhanced Administrative Co-operation in the Field of (Direct) Taxation, Member States agreed to EOIR, AEOI (later including the international AEOI Standard), spontaneous EOI and participation in administrative inquiries (EU Council Directive 2011/16/EU).11

Other initiatives have also been implemented allowing for AEOI. In 2010, the United States enacted the Foreign Account Tax Compliance Act (FATCA) requiring foreign financial institutions to perform specified due diligence procedures to identify and report information on foreign assets held by United States account holders, or be subject to a withholding tax of 30% of the gross amount of certain payments made from the United States to these institutions. Many countries and jurisdictions have implemented FATCA by entering into intergovernmental agreements (referred to as FATCA Intergovernmental Agreements, or IGAs) with the United States offering a bilateral approach to FATCA implementation under which the United States also provides certain (albeit more limited) information in return (OECD, 2014).12

A variety of legal means, therefore, exist to facilitate AEOI and many countries have engaged in AEOI with each other for a relatively long period of time. However, a key challenge has been standardising the information exchange across jurisdictions. In 2014 the OECD finalised the AEOI Standard for the automatic exchange of financial account information, developed in close co-operation with the G20, the EU and other stakeholders. This standard builds on the FATCA IGA to maximise efficiency and minimise costs. It is now a new global standard on tax transparency, having been endorsed by the Global Forum in 2014.

The AEOI framework requires Financial Institutions (FIs) to perform specified due diligence procedures to identify the financial assets and accounts which they hold on behalf of taxpayers from jurisdictions with which their tax administration exchanges information. The FIs are required to report information to tax administrations in the jurisdictions where they are located. The tax administrations then exchange that information with each other on an annual basis.

The CRS is seen as an effective way to tackle tax evasion and bank secrecy while minimising costs for governments and FIs. In the absence of a standardised approach, AEOI could be expected to impose higher costs on governments and FIs. It could also reduce the effectiveness of information exchange, as it would be more difficult to implement a standardised method of utilising the information for tax compliance purposes. In addition, the absence of a standard may create uncertainty among interested stakeholders on what information should be collected or exchanged.

International instruments require a separate agreement to operationalise AEOI. This is the case due to the fact that most international instruments, either bilateral or multilateral, only include a general provision enabling governments to exchange information without providing details on how the exchange will take place and exactly what information will be exchanged. In other words, signing the Convention or the TIEA does not allow for AEOI to take place by itself. Governments must agree on the distinctive features of the exchange through a separate agreement, such as the details as to what exactly is to be exchanged and the form in which the AEOI will take place. An example of such an agreement is the Multilateral Competent Authority Agreement on Automatic Exchange of Financial Account Information (CRS MCAA).13 The CRS MCAA includes the specific details of the information that will be exchanged between countries, and provides details on how the AEOI can happen. In other words, the CRS MCAA activates and operationalises the AEOI between jurisdictions by referring them to the CRS. To date 97 jurisdictions have signed the MCAA to implement AEOI.14

The implementation of AEOI in detail

The network of AEOI relationships amongst OECD/G20 has also expanded substantially in recent years. Most AEOI relationships between OECD/G20 countries and IFCs exist under the CRS, though there are also other AEOI relationships in the form of FACTA IGAs. While many countries began exchanging in 2017, a further large part of the expansion of AEOI will only commence by September 2018. The network will continue to expand in 2019 and 2020.

The implementation of AEOI requires a series of steps and agreements to ensure that the nature of information exchange and the means by which it is exchanged are consistent across all countries. This requires: rules on the collection and reporting of financial information by FIs; IT and administrative capabilities in order to receive and exchange the information; a legal instrument providing for information exchange between the jurisdictions; and measures to ensure the standards of confidentiality and data safeguards are met. The AEOI Standard therefore incorporates requirements in relation to each of these areas.

In 2014 the Global Forum adopted the AEOI Standard and put in place a commitment process to ensure its rapid implementation to create a level playing field. Following this commitment process more than 100 jurisdictions including almost all financial centres have committed to implement the CRS with exchanges starting in 2017 or 2018. To ensure completeness in coverage, the commitment includes a commitment to exchange information with “all interested appropriate partners”, which are all those partners wishing to receive information and which meet the expected standards on the confidentiality and proper use of the data. The Global Forum is now monitoring, assessing and assisting in the delivery of the commitments.

For AEOI to be effective there are four core requirements that must be met. These are:

  • Translating the due diligence and reporting rules into domestic law; as discussed, the detailed due diligence and reporting rules contained in the CRS must be translated into domestic law to ensure FIs perform the due diligence procedures and obtain and report the required information;

  • Selecting a legal basis for the AEOI to take place; countries will need to agree with other countries on the international instrument (bilateral or multilateral) under which the EOI will take place – in practice almost all jurisdictions have favoured the multilateral approach;

  • Putting in place IT and administrative infrastructure and resources; and

  • Protecting confidentiality and safeguarding data to guarantee that all collected and exchanged information will be kept confidential and that there are proper remedies that will be implemented in case of a breach.

Non-fulfilment of any of the requirements makes it impossible for AEOI under the CRS to take place.

Overview of the CRS

The CRS is the main part of the AEOI Standard as it contains the detailed rules and procedures that FIs must follow to ensure the relevant information is collected and reported. It provides for a standard on the due diligence procedures needed to identify who is a reportable person (RP), what are reportable financial accounts (RFAs), and what information needs to be reported. As has been mentioned, the benefit of adopting a common approach is to tackle tax evasion and bank secrecy with greater efficiency while minimising costs for governments and FIs.

Only Reporting Financial Institutions (RFIs) are required to perform the due diligence procedures outlined in the CRS. That is, only RFIs are required to report information in relation to certain individuals and entities to tax authorities if those individuals or entities hold a financial account with the RFI. The CRS defines RFIs as FIs located in a participating jurisdiction. The term FI includes depository institutions, custodial institutions, investment entities and specified insurance companies (see Annex C for more details).

Only entities can be considered RFIs.15 In other words, individuals directly engaging in any of the activities described above are not considered entities. Only entities can be considered to be FIs, and only those located in participating jurisdictions, may be a RFI provided they are not specifically exempted.

Some FIs are specifically excluded from these requirements because they pose a low risk of being used to evade taxes. The CRS includes a list of FIs that are specifically excluded. The list includes certain governmental entities, international organisations or central banks, broad and narrow participation retirement funds, pension funds of a governmental entity, qualified credit card issuers, exempt collective investment vehicles, certain trusts, and any other entity that presents a low risk of being used to evade taxes in the view of the signatory jurisdiction (see Annex C for further details).

RFIs must review the financial accounts they maintain to determine whether they are RFAs or not. A reportable account means an account maintained by FIs, and includes a depository account, a custodial account, equity and debt interests in certain investment entities, cash value insurance contracts, and annuity contracts (see Annex C for further details). The CRS specifically excludes certain retirement and pension accounts, non-retirement tax-favoured accounts, term life insurance contracts, estate accounts, escrow accounts, depository accounts due to unreturned overpayments, accounts established by court orders, accounts created for the sale, exchange or lease of real or personal property, and other low-risk accounts (see Annex C for further details). For financial accounts to be reportable accounts, they must be held by RPs.

  1. Only information about RFAs held by RPs will be subject to AEOI. Therefore, RFIs must identify whether the account holder is a RP. According to the CRS, a RP is any account holder who is a tax resident in a participating jurisdiction with which the jurisdiction where the RFA is located is exchanging information under the AEOI Standard. The following account holders, however, are not considered RPs:

  2. A corporation, the stock of which is regularly traded on one or more established securities markets, and any corporation that is related to this entity;

  3. A governmental entity;

  4. An international organisation;

  5. A central bank,

  6. A financial institution.

Table 5.3 lists the asset types that will be covered by the CRS.

Table 5.3. Asset types that will be covered by EOIR/AEOI

Type of Asset


Bank Deposits

Information will be exchanged as regards of commercial accounts, checking accounts, and, savings accounts.


An entity that owns bonds for the account of others is considered a custodial institution and thus, a reporting financial institution.

Equities (shares)

An entity that holds shares of stock in a corporation or partnership for the account of others is considered a custodial institution and thus, a reporting financial institution.

Interests in investment entities (including collective investment vehicles)

Entities that primarily conduct as a business investment activities or operations on behalf of other persons, and those that are managed by other financial institutions are considered investment entities and thus are potentially reporting financial institutions. Any debt or equity interest an individual or entity has in one of these vehicles may need to be reported.

Cash-value insurance contracts.

Information will be exchanged as regards cash value insurance contracts. These are insurance contracts that have a cash value. The term “cash value” is defined by the CRS as the greater of the amount that the policy holder is entitled to receive upon surrender or termination of the contract, and the amount the policyholder can borrow under or with regards to the contract.

Implementation of EOI

Exchange of information represents a “step-change” in transparency in the international tax environment, first with more widespread EOIR and then more recently with AEOI. As discussed above, the development of EOI, and in particular the advent of the CRS as a means to standardise EOI, is a significant new tool in the fight against tax evasion.16 However, continued efforts are required to ensure that the implementation of the CRS is as effective as possible. This section discusses these challenges and the efforts of the Global Forum and the OECD in this regard.

Administrative capacity

Collecting CRS information from FIs and exchanging it with participating tax administrations is only part of the implementation challenge. And while the deterrent effect in relation to the move to AEOI may be substantial, tax administrations also need to develop the capacity to use the information received in an efficient way to combat tax evasion. In this regard, work is ongoing within the Forum on Tax Administration to share knowledge and experience amongst tax administrations (see, for example, OECD (2016)) and the Global Forum is providing associated technical assistance.

Non-participating jurisdictions

Countries and jurisdictions not yet committed to exchanging information automatically under the CRS or any other standards will neither collect nor share any financial information as part of AEOI process. It is crucial that as many relevant jurisdictions as possible commit to AEOI to achieve effective EOI coverage and optimal international tax transparency. To address this, through the Global Forum members can identify “jurisdictions of relevance” for tax transparency. On identification by the Global Forum as a jurisdiction of relevance, countries can be asked to be members of the Global Forum or peer reviewed in absentia. Empirical evidence of the importance of the coverage of the EOI network is outlined in Section 5.4 below.

Non-reportable persons, accounts and assets

The potential for lack of coverage of non-reportable persons may generate challenges for the effective implementation of EOI. Only financial accounts held by RPs at a reporting financial institution will be reported under the CRS. In legal terms, a RP is any person (whether an individual, entity or legal arrangement) who is a tax resident in a reportable jurisdiction. However in some instances it may be challenging to ensure that the person who is in fact the beneficial owner of certain assets is identified. However, the current round of peer reviews with respect to EOIR now includes requirements relating to the access to and exchange of beneficial ownership information (OECD, 2016) and the future peer reviews in relation to AEOI will also need to incorporate these aspects.

Taxpayers may respond to more effective taxation of an asset by shifting savings to other forms of assets. Under the CRS, most financial accounts maintained by a financial institution are reportable (see Section 5.3). However, the CRS was always designed to focus on financial accounts, which is a particularly high-risk area. This therefore excludes a variety of kinds of assets which may include real assets as well as certain kinds of accounts that are non-reportable accounts. Example of real assets that are not currently reportable include:

  • Jewellery, art, cars and horses, although establishing effective AEOI over some of these assets might face additional challenges due to the difficulties in valuing and taxing these items, although EOIR regarding these assets is covered under the EOIR standard.

  • The direct ownership of commodities such as gold and silver, or less long-lasting ones such as rubber, live cattle and wheat, etc.

  • Real immovable property.

Continued vigilance will be required to ensure that these assets do not become an obstacle to effective international tax transparency as taxpayers may shift their assets held in financial products to these other types of assets.

In addition to some forms of assets that are not covered under the CRS, some financial accounts are also specifically treated as low risk accounts and excluded from reporting by the CRS. These include certain retirement and pension accounts, non-retirement tax-favoured accounts, term life insurance contracts, estate accounts, escrow accounts, depository accounts due to non-returned overpayments, accounts established by court orders, accounts created for the sale, exchange or lease of real or personal property, and other low-risk accounts (see Annex C for further details).

The Global Forum is reviewing all non-reportable accounts to make sure they are low risk. Jurisdictions themselves are also required to monitor low risk accounts to ensure they remain low risk. In addition, any scheme found to be used to hinder effective EOI can be disclosed through the CRS disclosure facility. This will allow these schemes to be systematically analysed by the OECD with a view to assessing the risk they present to the overall integrity and effectiveness of the CRS and to reach an agreement on appropriate courses of action (see Box 5.1).

Box 5.1. CRS Disclosure Facility

The CRS was designed with a broad scope in terms of the financial information to be reported, the Account Holders subject to reporting and the FIs required to report, in order to limit the opportunities for taxpayers to circumvent reporting. It also requires that jurisdictions, as part of their effective implementation of the Standard, put in place anti-abuse rules to prevent any practices intended to circumvent the reporting and due diligence procedures.

On 5 May 2017, the OECD launched a public disclosure facility for information on schemes designed to circumvent the application of the CRS.1 This online facility provides practitioners, financial advisers, and civil society with the means to disclose schemes that purport to circumvent the application of the CRS. Those with knowledge of such schemes are strongly encouraged to come forward and make use of this facility, including on an anonymous basis.

Using the Common Reporting Standard Disclosure Facility, interested parties are invited to describe the identified loophole, scheme, product, or arrangement that may be used for circumventing the CRS, and to provide as much supporting information as possible. Those who come forward are also allowed, if desired, to specify the countries or regions in which the scheme is being used, and to share any publicly available documents in which the scheme/product/arrangement is described/promoted.

1. The facility can be accessed through the Automatic Exchange Portal in the following link:

Source: OECD (2017c), The fight against offshore tax evasion continues: CRS disclosure facility delivers first results, .

Ensuring effective implementation

In order to maximise the effectiveness of the AEOI Standard and minimise the implementation challenges the Global Forum is responsible for monitoring the effective implementation of the AEOI Standard and ensuring any failings are addressed in order to deliver a level-playing field. This includes all aspects of AEOI implementation. Almost all jurisdictions have already been assessed in relation to the confidentiality and data safeguard standards they apply and the Global Forum is part-way through assessing the domestic legislative provisions implementing the AEOI Standard in each jurisdiction to ensure they properly reflect the detailed rules in the CRS and the associated Commentaries. There is also a process in place to identify and address any gaps in international exchange networks. A Common Transmission System has also been put in place to overcome some of the operational challenges and extensive technical assistance is being provided as necessary.

Given that exchanges have only just commenced amongst a subset of the jurisdictions it is not possible to assess effectiveness in implementation at this stage, although the Global Forum is currently developing a detailed framework for the peer reviews of the implementation of the AEOI Standard which are due to commence in 2019-2020. A major component of this will be ensuring FIs comply with their obligations under the AEOI Standard. The next important step will be the development of a robust Terms of Reference and methodology for the peer reviews of the effectiveness of the implementation of the AEOI Standard to ensure the AEOI Standard works as intended as a tool to tackle tax evasion.

5.4. The impacts of tax transparency

There have been substantial changes in the EOI environment over recent years. This means that a key policy question becomes how taxpayers will respond to these changes? Assessing what the taxation of foreign source income will look like is subject to considerable uncertainty. This section provides some preliminary evidence that EOI results in a reduction in incentives to hold wealth offshore. This section also identifies some of the potential limitations of current approaches to assessing the impact of more effective EOI.

Offshore wealth and the international capital environment have been a source of increased public debate in recent years. As a consequence, there is a small but expanding amount of empirical literature on this subject. This section will concentrate on those studies that specifically attempt to assess the impact of international information exchange on tax compliance and collections. There are, however, a wide variety of other papers attempting to characterise the size of hidden wealth as well as to assess taxpayer responses to international tax and to other international tax policies such as withholding taxes and blacklisting.

Efforts to evaluate the effectiveness of EOI rely on estimating the size of tax evasion in the first place, which by its nature is challenging and subject to significant uncertainty. For this reason many studies in this area do not focus on estimating the impact of EOI on tax evasion (which is difficult to estimate), but rather focus on a proxy for tax evasion: e.g. financial investments and bank liabilities in international financial centres. The assessment of the impact of EOI is uncertain, in part, because there are many reasons apart from tax evasion that may explain why assets may be held in foreign jurisdictions. Moreover, what constitutes an international financial centre is itself difficult to define. Much more research in this area is needed to effectively estimate the impact of EOI on international tax evasion and to develop tax policy recommendations for the future. In addition, because this literature is growing quickly, some of the papers cited in this section are unpublished. For all these reasons, the estimates cited in this section should be interpreted with caution.

Huizinga and Nicodème (2004) wrote one of the first papers to analyse the impact of information exchange. This paper uses data on bank liabilities from the Bank for International Settlements (BIS) to investigate the tax and non-tax factors that impact the holdings of bank liabilities by non-residents in a sample of 25 countries ending in 1999 (the year the Global Forum was set up). While they analyse a wide variety of explanatory variables from typical gravity models of trade and investment, the tax factors they investigate include withholding taxes and wealth taxes. They also examine whether international EOI impacts bank liabilities, although they do not make a distinction in their analysis between EOIR and AEOI. Most of the effects in the study likely stem from the impacts of EOIR, as the network of AEOI agreements was very limited at the time.

They find that information exchange has little impact on deposit holdings. However, their paper is hampered by two factors. First, while the paper relies on a time-series of data regarding bank liabilities, only one year of data is available on the information exchange network. This means that when analysing the impact of information exchange, it is not possible to control for country-specific factors such as withholding tax rates or GDP. Moreover, the information exchange data are for 1999. Out of 288 potential country-pair information exchange relationships in their dataset, only 67 entries show the presence of international information exchange. One-way exchanges are reciprocated in only 30 cases. As highlighted in Section 5.3, information exchange networks have expanded substantially since then following the G20’s call for an end to bank secrecy and the creation of the Global Forum to ensure the effective implementation of EOI worldwide. Moreover, as there was very little AEOI in 1999 it is difficult for the analysis to distinguish between the impact of EOIR and AEOI.

Huizinga and Nicodème argue that the limited impact of information exchange on bank liabilities in their sample could be a result of “the haphazard pattern of international information exchange at present”. They also suggest that “for [information exchange] to become more effective, the quality of the information exchanged may need to be improved, for instance through the adoption of a common protocol regarding tax identification numbers. Also, the international exchange of information has to cover most industrialized countries and other financial centres to be truly effective.” These two goals have largely been accomplished by the advent of the CRS and the work of the Global Forum in recent years.

Johannesen and Zucman (2014) study an expanded dataset from the BIS. This study incorporates data on foreign bank liabilities in 41 countries, and the time series runs from the fourth quarter of 2003 to the second quarter of 2011. This is a substantially larger dataset than Huizinga and Nicodème. Unlike Huizinga and Nicodème, they use a time-series dataset of EOI treaties which allows them to analyse the impact of the development of the EOI network over time, controlling for country-pair fixed effects, as well as time fixed effects. This can facilitate controls for events such as the financial crisis that impact bank liabilities in many countries at the same time.

Unlike the Huizinga and Nicodème paper, the Johannesen and Zucman study examines EOIR specifically. They suggest that EOIR does reduce bank liabilities in IFCs, by between 11% and 16% depending on the analytical specification. However, Johannesen and Zucman suggest that EOIR does not result in repatriation of foreign wealth to taxpayers’ countries of residence. They suggest that assets are shifted from country to country in response to the expansion of the EOIR network. This finding is based on the fact that in their estimation total global liabilities in IFCs have not fallen over the period they study. They argue that the lack of repatriation is evidence that tax is still being evaded on foreign wealth (though foreign wealth could remain in a foreign jurisdiction after having been declared to tax authorities). Based on this, these authors argue that “the G20 tax crackdown” had been ineffective.17 Although their data captures time series variation in EOI, in contrast to that of Huizinga and Nicodème, these authors do not analyse: (1) the impact of the expansion of the EOIR network after 2011; (2) the continued impact of the Global Forum’s peer review process which was a key driver in the increasing EOI networks; or (3) the widespread introduction of AEOI which began in 2017. This is because BIS data for this period was not available at that time.

Like Huizinga and Nicodème, Johannesen and Zucman highlight the fact that the EOI network is incomplete. They suggest that this means that when any given IFC signs and EOI agreement, tax evaders simply transfer funds from that jurisdiction to another jurisdiction that has not signed an EOI agreement. They say that “[a] comprehensive multilateral agreement would prevent tax evaders from transferring their funds from haven to haven”, alhough they also suggest that “even in the presence of a complete network of EOIR treaties, there may remain scope for improved tax collection by making treaties more demanding.”

In an unpublished paper, Gorea (2017) assesses the impact of tax treaties using equity and debt holdings of United States residents only.18 He uses data from 2002 to 2013, with the sample size of 43 IFCs. He finds that tax treaty signature has no effect on inflows in the sample. He does, however, find that treaty signature has a statistically significant positive impact on flows to IFCs that have not signed treaties with the United States. He supports the argument in Johannesen and Zucman that EOIR signature results in asset-shifting, and this highlights the importance of a securing a comprehensive network of agreements. He also finds larger effects for treaties that override bank secrecy laws; suggesting that this aspect of the development of EOI is an important factor in successful implementation.

The Gorea paper is different from both the Huizinga and Nicodème and Johannesen and Zucman papers in that he focuses on overall portfolio investment and not simply on bank liabilities. This means that he accounts for forms of tax evasion that may not be mediated through banks (for example investment in companies that hold immovable property). Like the Huizinga and Nicodème and Johannesen and Zucman papers, the analysis focuses on EOIR and not on AEOI – it does not account for the presence or absence of FATCA IGAs. The paper attempts to separately assess the impact of treaties that override bank secrecy provisions and those that do not. This highlights the importance of overriding bank secrecy in ensuring the EOIR instruments are as effective as possible.

Gorea focuses only on the impact of tax treaties on investment flows from the United States to IFCs. This means that he may miss a large amount of cross-national variation available with a larger data sample. He suggests that the incomplete nature of the United States tax treaty network could be a source of the statistically insignificant impact of tax treaties on investment flows, writing that rather than repatriating funds, investors simply choose to “relocate funds to other tax havens”. He also argues that “this study provides empirical support for the theory that … a multilateral treaty would have much greater effect on curtailing tax evasion, because it would limit asset relocations to jurisdictions with fewer information exchange agreements”.

In another unpublished paper, Omartian (2016) analyses the impact of signature of TIEA and commitment to implement the CRS on incorporation of companies in IFCs by foreign residents. His data sources include files released as part of the Panama Papers leak. He notes a decline in incorporation activity of the clients of the Panamanian law firm Mossack Fonseca in the years from 2012 to 2015. He also notes an increase in the terminations (i.e. liquidations and dissolutions) of companies held by Mossack Fonseca clients over this period, and that “terminations … accelerate at the tail end of the sample, eclipsing incorporations in 2014.” He suggests that the terminiations are evidence of the impact of EOI.

Analysing the impact of TIEA signature, he finds that “the number of incorporations per month for an investor country-jurisdiction pair falls roughly 12%” in the aftermath of a TIEA signature. When analysing the effects of the CRS on incorporation activity, he finds that incorporation activity occurring from investors in a country that publicly commited to exchange information fell by roughly 6 percentage points after the commitment.

Overall, the Omartian paper finds that commitments to the CRS and having TIEAs in place did have negative impacts on incorporations facilitated by Mossack Fonseca. However, it may not be the case that all of these incorporations were for the purpose of tax evasion, or that reducing incorporation levels are a good indicator of the EOI successfully tackling tax evasion and allowing for better taxation at the residence level. Nor does the paper assign a nominal figure to the tax revenue impact of the EOI. It is, however, one of the first papers to analyse the impact of commitments to the CRS, as opposed to information exchange efforts more broadly. He does not, however, test the impact of the implementation of the CRS itself, as the first exchanges under the CRS did not take place until September 2017.

A recently produced (and also unpublished) paper, Menkhoff and Miethe (2017), also considers these issues with new data. They use an expanded set of bilateral BIS data relative to the dataset used by Huizinga and Nicodème and by Johannesen and Zucman. This dataset contains data from 46 countries between 2003 to 2016. These authors use a more up-to-date set of EOI agreements than Johannsen and Zucman, and find a larger impact of EOIR agreements, which are found to reduce liabilities held by non-IFC residents in IFCs by 36%. This contrasts to the 11-16% found by Johannsen and Zucman.

In addition, Menkhoff and Miethe not only analyse inflows to IFCs but also outflows from these centres, based on the argument that investments in IFCs are subsequently “round-tripped” to other investment locations (Hanlon, Maydew, and Thornock, 2015). They find that EOI signature also reduces outflows from IFCs to non-IFCs in addition to inflows to IFCs from non-IFCs.

Overall, the literature provides suggestive evidence that EOI can be effective in reducing tax evasion through offshore locations. However, the literature is hampered by lack of data on offshore activity which is difficult to measure. The literature also provides evidence of factors that may limit the effectiveness of EOI. Huizinga and Nicodème suggest that EOI in 1999 has limited effects on bank liabilities due to the relatively incomplete nature of the EOI network at that time. Johannesen and Zucman find that EOIR is effective, but only in terms of reducing bank liabilities bilaterally, not in terms of reducing offshore bank liabilities overall. This highlights the extent to which early EOI efforts have been hampered by their patchwork nature. Given that the EOI network is continuing to expand, it is therefore likely that the impact of EOI has strengthened as the network has continued to expand. To assess this, it is important that the impact of EOIR and AEOI be regularly evaluated.

5.5. The implications of the expansion of EOI

In the last decade international progress in the field of tax transparency and EOI for tax purposes has been unprecedented. Co-operation between tax authorities has rapidly expanded and jurisdictions have been able to increase the effectiveness with which they tax their residents, especially on income resulting from assets held offshore.

Forecasting the impact of the expansion of the EOI network and the introduction of the CRS is challenging, in part because of the complex nature of estimating the size of global tax evasion to begin with (see for example, Pellegrini et. al. (2016) and Zucman (2013)). Existing studies suggest that EOI does reduce offshore liabilities, which means that it should enable more effective residence-based taxation.

It is also clear that the CRS substantially increases the chances of tax evasion being detected by authorities, which increases the risks that individuals engaging in tax evasion will be subjected to serious penalties and even possible prosecution. It also raises the legal, accounting and administrative costs of those wishing to persist with tax evasion. It does so by increasing the complexity of the schemes required to attempt to circumvent the CRS, and by increasing the likelihood that such attempts will fail. There is evidence to suggest that these increased costs can have a positive impact on compliance, as taxpayers tend to respond by reducing evasion and avoidance (Slemrod and Yitzhaki, 2002).

Many taxpayers will change their behaviour as the costs of attempting to continue to evade taxes become even higher and the risks of detection also increase. In addition, as most participating jurisdictions have put in place voluntary-disclosure programs, taxpayers have been provided with a strong incentive to come forward and start reporting their assets. Indeed, the evidence suggests that many have already done so: the most recent data suggest that countries have already received close to EUR 80 billion in unplanned additional revenue as a result of voluntary disclosure programmes and other similar initiatives in the lead-up to the first exchanges under the CRS (OECD, 2017b). Such voluntary disclosure programs are an integral part of a strategy to strengthen residence based taxation.

However, some taxpayers may accept a higher degree of risk as they assess the risks of their tax evasion activities becoming detected. In cases where the costs of implementing arrangements to remain outside the scope of the CRS/AEOI are lower than the taxes they would otherwise be liable to pay and where the complexity of the legal and administrative arrangements necessary to evade the tax are relatively small given the tax savings, taxpayers will have incentives to use mechanisms that ensure they remain outside the scope of AEOI.

As the costs and complexity associated with implementing arrangements to circumvent the CRS increase, it is likely that only those individuals that have the largest amounts of potential tax liability at stake and those that have a very high appetite for risk will be prepared to continue to arrange their affairs in ways that may conceal their assets with a view towards evading taxes.

The effectiveness of the CRS as a tool to enable more efficient taxation of capital income and assets will be limited by the extent to which taxpayers can and are willing to shift holdings of income and wealth to asset classes not covered by the CRS. Direct investment in real property is one such class. While, it may be possible that non-financial assets could be brought under the scope of an expanded AEOI standard at some future point, it should be noted that real-estate investments would still be subject to property taxes in the jurisdictions where the property is located. This raises the issue of the taxation of property covered elsewhere in this study, particularly given the relatively benign impacts of property taxation on economic growth (OECD, 2010).

Seminal papers in taxation have argued that the efficiency consequences of income taxation can be measured through the summary statistic of the elasticity of taxable income (Feldstein, 1999). If taxable income falls too much in response to taxation, then taxation becomes more costly. In respect of taxation of foreign source income and taxation on a residence basis, this elasticity has been high for many years. However, the core impact of the CRS from an economic perspective is that the strengthened EOI standards are likely to play an important role in reducing the elasticity of taxable income - especially capital income - with respect to the tax rate.

Recent papers have also highlighted the extent to which the elasticity is not a fixed factor for policymakers but rather itself an instrument of government policy that policymakers can reduce to varying degrees with the right policy mix (Slemrod and Kopczuk, 2002). By making it harder for taxpayers to conceal income and wealth in foreign jurisdictions, the expansion of EOI does exactly this. By making tax evasion more costly, it reduces one of the key elasticities pertaining to the taxation of top incomes (Piketty, Saez and Stantcheva, 2011). Holding other factors constant, the optimal response to a reduced elasticity from a policymakers’ perspective is an increase in the tax rate; in this case, with respect to capital income. These issues will be discussed further in Chapter 6.


Alstadsaeter, A., Johannesen, N., and Zucman, G. (2017), “Who owns the wealth in tax havens? Macro evidence and implications for global inequality”, NBER Working Paper, (23805), Cambridge, .

Arulampalam, W., Devereux, M., and Maffini, G. (2012), “The Direct Incidence of Corporate Income Tax on Wages”, European Economic Review, Vol. 56/6, pp. 1038-1054.

Brys, B., Perret, S., Thomas, A. and O’Reilly, P. (2016), “Tax Design for Inclusive Economic Growth”, OECD Taxation Working Papers, No. 26, OECD Publishing, Paris, .

Diamond, P., and Mirrlees, J. (1971), “Optimal Taxation and Public Production I: Production Efficiency”, The American Economic Review, Vol. 61/1, American Economic Association, pp. 8-27.

Feldstein, M. (1999), “Tax avoidance and the deadweight loss of the income tax”, Review of Economics and Statistics, Vol. 81/4, The MIT Press, pp. 674-680.

Financial Action Task Force (2017), FATF website, .

Fuest, C., Peichl, A., and Siegloch, S. (2018), “Do Higher Corporate Taxes Reduce Wages ? Micro Evidence from Germany”, TheAmerican Economic Review. (forthcoming)

G20 (2009), London Summit Leader’s Statement, OECD and the G20, .

Global Forum on Transparency and Exchange of Information for Tax Purposes (2014), Final Report to the G20 Development Working Group Automatic Exchange of Information: A roadmap for developing country participation, Paris, .

Gorea, D. (2017), “Catch Me If You Can: U. S. Tax Treaties and Shifts in Offshore Investments”. Working Paper, European University Institute.

Hanlon, M., Maydew, E., and Thornock, J. (2015), “Taking the long way home: US tax evasion and offshore investments in US equity and debt markets”, The Journal of Finance, Vol. 70/1, The Journal of the American Finance Association, pp. 257-287.

Huizinga, H., and Nicodème, G. (2004), “Are international deposits tax-driven”, Journal of Public Economics, Vol. 88/6, Elsevier, pp. 1093-1118.

Johannesen, N., and Zucman, G. (2014), “The end of bank secrecy? An evaluation of the G20 tax haven crackdown”, American Economic Journal: Economic Policy, Vol. 6/1 B, American Economic Association, pp. 65-91.

Keen, M., and Ligthart, J. (2006), “Information sharing and international taxation: A primer”, International Tax and Public Finance, Vol. 13/1, pp. 81-110

Menkhoff, L., and Miethe, J. (2017), “Dirty money coming home: Capital flows into and out of tax havens”, Discussion Papers, No. 1711, DIW, Berlin.

Milanez, A. (2017), “Legal tax liability, legal remittance responsibilityand tax incidence: Three dimensions of business taxation”, OECD Taxation Working Papers, No. 32, OECD Publishing, Paris, .

OECD (2017a), Fixing Globalisation: Time to Make it Work for All, OECD Publishing, Paris, .

OECD (2017b), OECD Secretary-General’s Report to G20 Finane Ministers, Baden-Baden, Germany, .

OECD (2017c), The fight against offshore tax evasion continues: CRS disclosure facility delivers first results,

OECD (2017d), Standard for Automatic Exchange of Financial Account Information in Tax Matters, Second Edition, OECD Publishing, Paris, .

OECD (2016), Exchange of information on request: handbook for peer reviews 2016-2020, OECD Publishing, Paris.

OECD (2014), “Commentaries on the Common Reporting Standard”, in Standard for Automatic Exchange of Financial Account Information in Tax Matters, OECD Publishing, Paris, .

OECD (2014), Standard for Automatic Exchange of Financial Information in Tax Matters, OECD Publishing, Paris, .

OECD (2010), Tax Policy Reform and Economic Growth, OECD Publishing, Paris, .

OECD (2006), Taxation of Capital Gains of Individuals: Policy Considerations and Approaches, OECD Publishing, Paris, .

Omartian, J. (2016), “Tax Information Exchange and Offshore Entities: Evidence from the Panama Papers”, SSRN Electronic Journal, 10/2139.

Pellegrini, V., Sanelli, A., and Tosti, E. (2016), “What do external statistics tell us about undeclared assets held abroad and tax evasion?”, Questioni de Economia e Finanza Occasional Papers, No. 367, Banca d’Italia.

Piketty, T., Saez, E., and Stantcheva, S. (2011), “Optimal Taxation of Top Labor Incomes: A Tale of Three Elasticities”, American Economic Journal: Economic Policy, Vol. 6/1, The National Bureau of Economic Research, pp. 230-271.

Sheppard, Lee A. (2017), “Is bitcoin going out of style?”, Tax Notes International, , (accessed 11 September 2017).

Slemrod, J., and Kopczuk, W. (2002), “The Optimal Elasticity of Taxable Income”, Journal of Public Economics, Vol. 84/1, Elsevier, pp. 91-112.

Slemrod, J., and Yitzhaki, S. (2002), “Tax Avoidance, Evasion, and Administration”, in Handbook of Public Economics, Vol. 3, Elsevier, pp. 1423-1470.

Zucman, G. (2013), “The missing wealth of nations: Are Europe and the U.S. net debtors or net creditors?”, The Quarterly Journal of Economics, Vol. 128/3, Oxford Academic, pp.1321-1364.


← 1.

← 2. See for example Chamley (1986), Judd (1985) and Atkinson and Stiglitz (1976).

← 3. This data is current as of January 2018.

← 4. As of July 2017, the Convention has been signed by 112 jurisdictions. Other instruments that are used to enable cross-border EOI between tax authorities include double tax treaties and bilateral tax information exchange agreements. The number of these instruments has also been steadily growing.

← 5. Different procedures may be required under the domestic laws of each country for an international agreement to be executed and implemented. In either case, international agreements are required for countries to have the legal grounds under which to exchange information among each other.

← 6. This list is based on that found in IMF (2004) and includes: Andorra, Anguilla, Antigua and Barbuda, Aruba, Bahamas, Bahrain, Barbados, Belize, Bermuda, Cayman Islands, Cook Islands, Costa Rica, Curaçao, Cyprus, Gibraltar, Guernsey, Hong Kong, Isle of Man, Jersey, Lebanon, Liechtenstein, Macao, Malaysia, Malta, Marshall Islands, Mauritius, Monaco, Nauru, Niue, Panama, Sint Maarten, Saint Kitts and Nevis, Saint Lucia, Saint Vincent and the Grenadines, Samoa, Seychelles, Singapore, Turks and Caicos Islands, Vanuatu, and the British Virgin Islands.

← 7. The term competent authority is the authority designated by each jurisdiction in charge of collecting and providing the requested information to the requesting jurisdiction.

← 8. Global Forum on Transparency and Exchange of Information for Tax Purposes, Exchange of Information on Request, (accessed 19 September 2017).

← 9. These are considered primary authoritative sources of AEOI which must be complemented by a number of secondary documents, such as a competent authority agreement, which give elements of context for the understanding, interpretation and implementation of AEOI.

← 10. The Directive was repealed on 10 November 2015 by the European Council as part of a tax transparency package presented by the European Commission to avoid a significant overlap that had developed with other European legislation.

← 11. In December 2014, the European Council adopted the Directive 2014/107/EU extending the scope of the EOI to include interest, dividends, and other types of income requiring Member States to implement reporting and due diligence rules fully consistent with those set out in the OECD CRS.

← 12. In 2012 the five largest European countries (France, Germany, Italy, Spain, and the United Kingdom) agreed with the United States on a reciprocal exchange of FATCA information under Intergovernmental Agreements (IGAs) concluded between the United States and each of the five countries.

← 13. Other MCAA agreements also exist, such as the Multilateral Competent Authority Agreement on Country by Country Reporting (CbC MCAA). Bilateral Competent Authority Agreements also exist.

← 14. This data is current as of 13 December 2017;

← 15. The term ‘entities’ being used throughout this document corresponds to the definition provided by the CRS, which includes only legal entities and legal arrangements.

← 16. While other forms and legal bases for EOI exist and are effective, the discussion in this section will focus on the CRS as it is the central multilateral mechanism to exchange information.

← 17. The renewed efforts to end bank secrecy and expand EOI after the April 2009 G20 meeting in London, UK.

← 18. The data used in this paper come from the Treasury International Capital (TIC) reporting system maintained by the U.S. Department of the Treasury (Gorea, 2015).