UK aid’s contribution to tackling tax avoidance and evasion

ICAI Score

In recent years, the G20 and the OECD have led a series of reforms on international tax, designed to help in the fight against international tax avoidance and evasion. DFID set out to influence these reforms so that developing countries could benefit from them. It also provided capacity building to help with implementing new tax standards.

DFID has helped developing countries to participate in various international tax processes; however, their ability to influence the content of the new standards was limited, and the international tax reform agenda failed to address a number of important issues for DFID’s partner countries.

DFID’s capacity building programmes on international tax have achieved some early results. Despite some initial problems, its partnership with HMRC is potentially a good model for collaborating on capacity building. However, various stakeholders expressed doubts that technical assistance on highly specialised international tax issues would have much impact, given the more basic problems with national tax systems. There are also concerns that the benefits to DFID’s partner countries of implementing the new standards may have been oversold.

We found that DFID does not have a clear approach to promoting ‘policy coherence for development’ in the tax area. It has not assessed areas of potential tension between UK tax policies and the needs of developing countries, nor made explicit decisions as to which issues to raise in cross-government dialogue.

We have given DFID’s work on international tax an amber-red score for relevance, because it is not well grounded in the needs and priorities of DFID’s partner countries. We awarded a green-amber score for effectiveness, in recognition of good cross-government collaboration and some early results on capacity building. An amber-red score for learning reflects that DFID has not made sufficient use of research or learning from its in-country tax programmes to inform its approach. Overall, this presents an amber-red score.

Nonetheless, we recognise that influencing international systems on matters such as tax is an important new frontier for the UK aid programme. We set out some recommendations to support future DFID and UK government efforts in this area.

Executive Summary

International tax evasion and avoidance is a global challenge, presenting significant costs for both developing and wealthy countries. In recent years, the G20 group of countries and the Organisation for Economic Cooperation and Development (OECD) – with the support of the UK government – have led a series of reforms to the international tax system to tackle the problem.

Working under the leadership of the Treasury, DFID sought to influence these international tax processes so that they could benefit developing countries, enabling them to mobilise more tax revenues for national development. While DFID did not set explicit objectives for its engagement with these processes, we identified two main objectives from its internal documents and our interviews with the tax team, against which to measure its performance:

i. To ensure that the needs of developing countries were considered in the G20 standard-setting processes.

ii. To ensure that developing countries received sufficient capacity building support to implement the agreed standards.

In this review, we assess DFID’s influencing work and capacity building on international tax, along with a number of other centrally managed programmes in the broader area of tax and development. DFID’s support for the G20-led tax processes has involved engagement by advisors in its central tax team and two main programmes: the Tax Transparency Programme (£7 million, 2014-17) and the HMRC Tax Expert Unit (£1.8 million, 2015-18). Its broader tax and development work includes support for the HMRC Capacity Building Unit (£22.9 million, 2014-24), which provides HMRC experts to support DFID bilateral tax programmes. These are included in the review to enable us to explore cross-government collaboration on capacity building.

This is a relatively small area of expenditure for DFID, with total commitments of £38.9 million over fourteen years, from 2010 to 2024. By comparison, DFID spent £32.6 million on in-country tax programmes in 2015 alone. However, international tax is one of the priority areas identified in the UK Aid Strategy for working ‘beyond aid’, by promoting changes to international systems that create more opportunities for developing countries[1] . This is an important new frontier for the aid programme. It also offers a timely opportunity to assess how well DFID works with other government departments.

We have designated this a learning review because it examines a relatively new area of activity for DFID, where learning on how to achieve results is at an early stage.

DFID had only limited success in making international tax processes more inclusive

DFID used a number of channels to pursue its influencing goals on international tax. These include working closely with the Treasury to feed into G8/G7[2] and G20 processes, and representing the UK government in the G20 Development and Anti-Corruption Working Groups. Feedback from other participants indicates that DFID played a prominent and constructive role in these processes, building its profile as a leading donor on tax.

DFID’s efforts to make the international standard-setting processes more inclusive of developing countries were, nonetheless, only partially successful. DFID supported the participation of developing countries in various G20 and OECD processes, including the OECD Tax and Development Task Force and the Global Forum on Transparency and Exchange of Information. However, key stakeholders from both OECD and developing countries agree that developing countries gained little practical influence over the new standards. Consultations with developing countries occurred late in the process, when the priorities had already been agreed.

We were only able to find evidence of two measures introduced in response to developing country concerns.[3] According to our survey and interviews, a number of issues of interest to DFID’s partner countries remain excluded.[4]

DFID leads the campaign for more donor support for capacity building although this is yet to produce significant results

One of DFID’s influencing objectives for the 2013 G8 summit at Lough Erne was to secure more donor funding for capacity building on international tax. Our review of the correspondence leading up to the summit showed that DFID secured references to capacity building in the G8 communiqué, but no commitment to increasing donor support. DFID itself remains the largest donor in the international tax arena.

DFID also supported the launch of the Addis Tax Initiative at the UN Financing for Development Conference in July 2015. This included a commitment by donor signatories to ‘collectively’ double their assistance (a formulation that provides flexibility to the responsibilities of individual donors). It is too early to say whether this commitment will be met. With DFID’s encouragement, the OECD body responsible for aid statistics is just beginning to monitor donor expenditure on tax.

The Addis Tax Initiative was developed by DFID and other donor countries with only limited consultation with developing countries and no explicit assessment of their needs. As a result, DFID and its donor partners had to lobby their partner countries at the Addis conference to sign up to the initiative. So far, only nine DFID priority countries have done so.[5]

DFID has provided some effective capacity building support, but it remains unclear whether partner countries will be able to implement the new standards

DFID supports implementation of the new tax standards through a number of centrally-managed programmes. These are beginning to deliver some early results. Various participating countries have adopted new regulations, signed international or bilateral agreements, developed action plans or established new specialist units within their tax administrations. Two of DFID’s partner countries, Kenya and Zambia, have successfully resolved some individual transfer pricing cases, albeit with the direct support of international experts.

However, some of the stakeholders we interviewed in our case study countries, including DFID advisors, expressed doubts that technical assistance on highly specialised international tax issues would have much impact, given more basic capacity constraints in national tax systems. There are also concerns that the DFID-supported OECD has taken an overly technical approach to capacity building, which fails to take sufficient account of the state of national tax systems.

DFID’s capacity building support on international tax was designed to be demand-led. Yet according to DFID’s own programme documents, demand from partner countries has been mixed. DFID and the OECD have responded by raising awareness of these issues in order to persuade African leaders of the benefits of implementing the new tax standards.

Despite these efforts, a number of stakeholders expressed their concern to us that these benefits may have been oversold. It is unclear whether measures such as exchange of information will prove effective, given more basic problems such as corruption in national tax administrations and the lack of effective sanctions and asset recovery mechanisms. Many developing countries grant overgenerous tax incentives to multinational corporations in order to attract investors. If multinational corporations have little tax liability in the first place, it is unlikely that measures to prevent profit shifting will generate significant additional revenue.

As a result, we are concerned that DFID’s influencing approach and its capacity building support on international tax have not been based on clear analyses of developing country needs, or on how international tax initiatives should be sequenced with broader reforms to domestic tax systems.

DFID has worked well with other UK government departments on influencing and has developed a promising model for collaboration on capacity building

DFID worked well with Treasury and HMRC to shape UK positions to take into G7/G8 and G20 processes. We saw evidence of DFID’s technical inputs being utilised by other departments, and that strong cross-government collaboration had contributed to effective UK advocacy.

In response to encouragement from the International Development Committee (IDC),[6]  DFID provided funding to HMRC to establish a Capacity Building Unit on tax. This is now funded from HMRC’s own aid budget, with estimated funding of £22.9 million over ten years. Through this unit, experts from HMRC are deployed to support DFID bilateral tax programmes for both short and long terms. While at an early stage, our case studies suggest that this is a promising model of collaboration. It enables DFID programmes to call on specialist expertise from HMRC, while ensuring that HMRC’s technical assistance is anchored in a broader strategy aligned to each country’s needs and priorities.

Beyond the Capacity Building Unit, we found that other departments do not look to DFID centrally for advice on how to provide effective capacity building. Nor is there much evidence of DFID trying to cultivate this role. DFID has now recognised that it needs to make a more active contribution on this issue.

There are value for money concerns with aspects of DFID’s capacity building support

In its international tax work, DFID has made efficient use of small investments of financial and human resources. It has combined its spending and influencing activities well, worked productively with other departments and used aid resources strategically to build up the capacity of its implementing partners.

However HMRC’s early use of funds was not good value for money. In the first year of the Capacity Building Unit, £1.17 million was spent on training HMRC tax experts for domestic roles in order to release existing staff for deployment abroad.[7] HMRC has never achieved more than half of its planned deployment. This highlights a key value for money risk as departments take on new aid delivery roles. It will take time for them to establish the systems, capacity and programmes needed to spend aid effectively in new country contexts.

There are also value for money concerns about short-term technical assistance. In our survey and country case studies, national tax authority officials expressed doubts that short-term, one-off missions added much value. Some DFID in-country tax advisors also expressed views that revenue authorities in many developing countries lack the capacity to absorb highly specialised, short-term inputs. Given the commitment in the Aid Strategy to making more use of the technical skills available across the UK government for aid delivery, this is an important finding. While counterparts appreciate the opportunity to interact with their UK peers, there is still much to be learnt about how to use peer-to-peer assistance to best effect.

DFID does not have a clear approach to promoting policy coherence for development in tax

Under the Busan Partnership for Effective Development Cooperation,[8] the UK is committed to pursuing ‘policy coherence for development’ – that is, promoting greater coherence across its public policies so as to increase opportunities for developing countries. Both the IDC[9] and the OECD[10] have encouraged DFID to do more in this area.

Our evidence suggests that DFID has not actively pursued policy coherence for development in the tax arena. A number of policy positions taken by the UK government, including a strong commitment to tax transparency, have been helpful to developing countries. However, the literature suggests that there are potential areas of tension between UK policies and developing country interests, including those concerning international tax competition and bilateral tax treaties. The International Monetary Fund (IMF) and the OECD have both advocated that developed countries undertake ‘spillover analysis’, to assess the indirect impacts of their tax policies on developing countries. The UK government has declined to do so, following the Treasury’s assessment that it would be impractical and time-consuming.

We take no view on the merits of UK policy position on ‘spillover analysis’ nor on tax more generally. We also recognise that, once a UK policy decision is made, it is binding on DFID. However, given that tax is highlighted in the Aid Strategy as a ‘beyond aid’ priority area, we would have expected to see evidence of DFID taking active steps to understand the impact of UK tax policies on developing countries and making informed judgments about raising areas of tension in cross-government dialogue. We found no evidence that this had been done.

DFID has made some useful investments in research, but has not drawn on evidence and learning effectively in its international tax work

There is a lack of solid evidence on how international tax issues affect developing countries, which hampers policy making. DFID has supported a research centre on tax and development at the University of Sussex by contributing £3.5 million (2010-16). Its research is well regarded; however in the absence of robust primary data from developing countries, its work on international tax remains largely theoretical.

We found little evidence of DFID’s tax team drawing on available research while setting its own priorities, in its dialogue across the UK government, or with external partners. Nor is there much evidence of DFID using the knowledge available within its country offices to inform its international influencing activities or central programmes on tax.

We also found that DFID did not establish explicit objectives for its influencing work on international tax, or any method of assessing its impact. In the absence of robust monitoring arrangements, we observed a tendency in some documents to overstate the results of its influencing efforts. If DFID hopes to achieve real impact though its ‘beyond aid’ engagements, it needs to adopt a more strategic approach to achieving and measuring results.

Conclusions and recommendations

DFID positioned itself well to take advantage of the opportunity presented by the G20 tax reform agenda to help developing countries benefit from the new international tax standards. It has made efficient use of limited advisory resources and combined influencing activities and aid programmes in complementary ways. It has demonstrated good cross-government collaboration and has achieved some positive early results with its capacity building, although there is reason to question the value for money of some of HMRC’s early work and of its one-off, short-term technical assistance. In light of these successes, we have given DFID a green-amber score for effectiveness.

However DFID’s influencing approach was not based on sufficient consultation with, or analysis of the needs of, its partner countries. While DFID supported developing country participation in certain international tax processes, their influence over the content of the new standards proved to be limited. DFID has also not pursued an active approach to promoting policy coherence on tax and development across the UK government. In light of these shortcomings, we have given DFID an amber-red score for relevance.

Finally, DFID has made some worthwhile investments in research in an attempt to address evidence gaps. Despite this, we found relatively little indication that DFID had used research or experience from its country programmes to support its influencing activities or to push the boundaries of policy dialogue on international tax. A lack of clear objectives, explicit strategy and monitoring arrangements for its influencing has also limited DFID’s ability to learn lessons and has led to over-optimistic assessments of its results. We have therefore given DFID an amber-red score for learning.

In light of the two amber-red scores for relevance and learning, and the green-amber score for effectiveness, we have given DFID an overall score of amber-red for its international tax work.

We have made a number of recommendations to improve future efforts in this area.

Recommendation 1: Learning on international tax issues

DFID should make better use of its in-country work on tax and anti-corruption to inform its influencing efforts and prioritise its programming on international tax.

Recommendation 2: Cross-government working

DFID should be more proactive in ensuring that other departments engaging in capacity building on tax and more generally, are able to draw on its experience of effective capacity building approaches and its knowledge of country contexts. This will require closer collaboration between departments, both at headquarters and in country.

Recommendation 3: A strategic approach to influencing

DFID should adopt a more systematic approach to influencing and cross-government working on international tax, with a stronger strategy and explicit objectives that are adequately resourced and properly monitored.

Recommendation 4: Policy coherence for development
DFID should take a more active approach to promoting policy coherence for development on international tax, for example by assessing the impact of UK tax policies and practices on developing countries and deciding whether to raise any points of tension in cross-government dialogue.

We have also proposed a number of ‘learning frontiers’ relevant to ‘beyond aid’ engagements by DFID and other government departments. These include capitalising on DFID’s expertise more generally and introducing a more active approach to policy coherence for development.

Introduction

Purpose of the review
In an increasingly globalised world, many wealthy individuals and multinational corporations have developed sophisticated cross-border strategies to avoid paying tax, using loopholes in the global financial system. The resulting loss of revenues affects both wealthy and poor countries. Tax havens and non-cooperative jurisdictions enable wealthy individuals to hide their wealth, evade tax and
launder money from illicit activities. There is heightened public interest in these issues following the recent leak of documents from the Panamanian law firm, Mossack Fonseca & Co. (see Box 2).

In its Aid Strategy, the UK government has committed itself to going ‘beyond aid’ in its pursuit of global poverty reduction[11] – that is, not just funding development programmes, but also working to change UK and global policies and systems to create more opportunities for developing countries. One of its priority areas for working beyond aid is tax. In recent years, DFID has worked through the G20 group of countries and the Organisation for Economic Cooperation and Development (OECD)[12]  to help developing countries implement new international standards designed to combat tax avoidance and evasion.

In this review, we explore DFID’s contribution to the G20-led tax reform agenda, which began in 2009 and is still underway. With support from the OECD, the G20 has developed a number of global standards and processes to promote greater transparency and tackle abusive forms of tax avoidance.

These include:

• Exchange of information on taxpayers between national authorities (including automatic exchange of information).

• Disclosure of who ultimately benefits from companies, trusts and other legal entities.

• Measures to limit the ability of multinational companies to avoid taxes through the way they manage their trade and investment flows.

These initiatives are described in more detail in the background section. A glossary of technical terms is provided in Annex 2.

This review explores how well DFID has used its influence across the UK government and with international partners to help developing countries benefit from these standards. This is the first external review of DFID’s engagement with the G20. It is also the first ICAI review of DFID’s approach to ‘policy coherence for development’ – that is, its efforts to identify and address potential conflicts between the UK’s international development objectives and other UK policy agendas.[13]

This review also assesses how DFID has used its aid programmes to promote the participation of developing countries in international tax standards, and to support other tax and development initiatives. Through a number of centrally managed programmes, DFID provides funding to HM Revenue and Customs (HMRC), the OECD and other partners for capacity building support.

The collaboration with HMRC and Treasury aligns with a UK government commitment under the Aid Strategy[14] to make better use of the complementary skills available across government in the delivery of UK aid.[15]

We have chosen to undertake a learning review because working beyond aid on international rules and standards is a relatively new activity for DFID. Particular emphasis is given to the way DFID has generated and applied evidence and learning to support its approach. As well as reviewing DFID’s performance to date, we draw attention to areas where DFID and other government departments could focus their learning in the future. We call these ‘learning frontiers’.

Scope of the review

The review covers three main areas of the DFID Financial Accountability and Anti-Corruption Team’s work on international tax.

• First, we examine DFID’s efforts to influence and support the international tax reform agenda.

• Second, we review other elements of DFID’s ‘tax and development’ agenda, including its contribution to the HMRC Capacity Building Unit (£22.9 million, 2014-24)[16]  and the Addis Tax Initiative (see Box 8).

• Third, we explore the link between these centrally managed initiatives and DFID’s country-level programming on tax, looking at whether they are coordinated, reinforce each other and enhance learning. (We have not reviewed DFID’s country-level tax programmes directly).

The total financial commitment to the initiatives covered by this review is £38.9 million over a fourteen year period (2010-24).[17]  This is a small investment by DFID’s standards, compared to the £32.6 million it spent on in-country tax programmes in 2015.

Table 1: Our Review Questions

Methodology

The methodology for this review involved four main elements:

i. An assessment of DFID’s centrally managed programmes and non-spending activities in relation to international tax. This included assessing DFID’s engagement with international processes, its work with other UK government departments, and the effectiveness and value for money of DFID’s programmes and activities. Our evidence comes mainly from stakeholder interviews and documents from DFID and other UK government departments.

ii. Four country case studies (Ghana, Ethiopia, Tanzania and Pakistan),[19]  which examine DFID’s programmes and activities from the perspective of particular countries. The case studies were prepared by reviewing documents and conducting telephone interviews with stakeholders, including national tax authorities, but did not involve country visits.

iii. A survey of DFID lead advisors on tax in country offices. 18 responses were received, which was a satisfactory 62% response rate.

iv. An extensive literature review and two roundtables with UK-based non-governmental organisations (NGOs) and private-sector representatives, to explore (i) the scale and impact of tax evasion and avoidance in developing countries; (ii) the main achievements and limitations of the international response and the UK’s position on international tax; and (iii) progress by signatory countries (including the UK) on international commitments. We also produced an annotated bibliography of analysis by southern organisations, to help capture developing country perspectives.

One of the challenges for this review was that DFID does not have explicit objectives for all of its international tax work. We conducted our own analysis of DFID’s objectives from documents and interviews, and used these to measure effectiveness.

Background

The G20 international tax reform agenda

The G20 is tackling tax evasion and avoidance by promoting transparency and cooperation

Since the 2008 global financial crisis, the G20 group of countries has emerged as the leading international forum on the global economy. In a series of declarations from 2009 onwards, the G20 has pledged to increase transparency and improve international cooperation in order to combat tax avoidance and evasion. Commitments were also made during the 2013 G8 summit in Lough Erne.

The G20 international tax agenda comprises three main initiatives: exchange of information, beneficial ownership, and base erosion and profit shifting (BEPS) (Figure 3). With the OECD Global Forum taking the lead, it has established standards for sharing information on tax payers between national tax authorities – initially on request and then from 2014 through an automated system. There is a peer review mechanism for monitoring countries’ compliance with these standards.

From 2013, the G8 countries agreed to begin sharing information on beneficial ownership – that is, on the individuals who ultimately stand to benefit from shell companies, trusts and other legal entities. This was part of a campaign against corruption and money laundering, as well as against global tax evasion. The G20 followed suit in 2014. This information will be shared between competent authorities under the new exchange of information standards.

The third initiative is a package of measures to fight international tax avoidance. Some multinational companies avoid tax by shifting profits between related legal entities in different jurisdictions – for example, through manipulating ‘transfer pricing’ of transactions between subsidiaries.[20]  The agreed mechanism for tackling this issue is the BEPS package of measures, endorsed by the G20 in Antalya in 2015. The package includes minimum standards, common approaches and best practices.

Developing countries have been invited into G20 international tax processes

The G20’s tax agenda was designed to address the needs and priorities of G20 countries. In an increasingly globalised world, these initiatives need to be global in reach to be fully effective. The G20 and its partner organisations have therefore encouraged developing countries to adopt the new international standards. In their communiqués, they have invited them to join the OECD’s Global Forum, and pledged capacity building support to help them implement the agreed standards.

Today 134 countries, including nine DFID priority countries, are members of the Global Forum.[21]  In early 2016, the OECD also invited developing countries to participate in its Committee on Fiscal Affairs, which oversees the BEPS initiative.

The UK position

The UK has played a leading role in the G20 tax agenda

Working closely with Germany, France and the US, the UK government has played a significant role in promoting tax transparency, particularly during the UK presidency of the G8 in 2013. The UK was the first country formally to commit to implementing country-by-country reporting. In 2015, it adopted legislation establishing a public register of beneficial ownership (excluding trusts), which goes beyond the agreed international standards. An Anti-Corruption Summit in London in May 2016 provided an opportunity for other countries to make commitments to publicising beneficial ownership.

Treasury is the lead department on the UK’s international tax policy, with support from DFID and other departments. To support this work, DFID has scaled up its centrally managed programmes on tax and development, from less than £5 million in 2012 to just below £40 million today (see Box 4).

Findings: How well has DFID contributed to the development of international tax standards in a way that benefits developing countries?

Tax policy is one of a number of areas where DFID has undertaken to work ‘beyond aid’ to defeat poverty and promote global prosperity.[22] DFID recognised that developing countries would need technical assistance from donors to implement the new international tax standards. It also saw an opportunity to influence the standards so as to enhance the benefits to developing countries. These objectives were later reflected in the UK Aid Strategy, in the form of commitments to tackling tax evasion and avoidance and allowing developing countries full access to automatic exchange of information.[23]

DFID did not set out explicit influencing objectives for its work on international tax. Through our interviews with DFID staff and our review of DFID documentation and correspondence in the lead-up to key events, we identified two underlying objectives:

i. To ensure that the needs of developing countries were considered in the standard-setting processes.

ii. To ensure that they received sufficient capacity building support to implement the agreed standards.

We assessed DFID’s progress against both objectives.

DFID used a range of influencing channels and aid investments to pursue its objectives

DFID used a number of channels to influence the international tax agenda. Its central team worked closely with the Treasury to feed into the 2013 G8 summit and the G20 Finance Track. It represented the UK in the G20 Development and Anti-Corruption Working Groups, taking part in the discussions and sharing briefings on UK positions (see Box 6). It also took advantage of the preparations leading up to the 2015 UN Financing for Development conference to promote the Addis Tax Initiative, alongside other donor countries and the host country, Ethiopia.

Through these efforts, DFID has been highly visible in discussions on the G20 tax agenda. The donor representatives we interviewed were in agreement that DFID played a prominent role. They also gave positive feedback on DFID’s role as co-chair of the Domestic Resource Mobilisation pillar of the G20 Development Working Group. Some linked the quality of DFID’s technical inputs to the specialist knowledge of its tax advisors in London, others to its substantial portfolio of bilateral tax programmes.

Its efforts to make the G20 processes inclusive had only limited success

DFID set out to influence the international tax processes so that the needs of developing countries were considered in the new standards. According to one internal document, its aim was to make the new standards ‘truly inclusive’.

Working with the OECD and other donor countries, DFID supported the participation of developing countries in various G20 processes, initially in the OECD Tax and Development Task Force[24] and the Global Forum, then later in various diagnostic exercises.[25] When the G20 agreed to a more structured dialogue with developing countries on the BEPS package, DFID responded with additional funding to enable the OECD Tax and Development Programme to support their participation.

One of the barriers to inclusive dialogue on international tax is that few low income countries have the resources or expertise to participate substantively in what are often highly technical discussions. DFID’s investment in the African Tax Administration Forum has helped to overcome this barrier, even if that was not its initial purpose. Set up in 2009 as a platform for cooperation among African tax authorities, it has proved to be a useful structure for representing African interests in international processes.

With financial support from DFID and other donors, the Forum has co-hosted regional events on international tax with the OECD. Various African countries, including three of our case study countries, have used the Forum to represent them in international dialogue.

Despite these efforts, the key stakeholders we interviewed from both OECD and developing countries agree that developing countries gained little practical influence over the international tax standards. In 2014, the OECD’s Development Working Group acknowledged that a number of issues important to developing countries – such as wasteful tax incentives, the lack of comparable data on transfer pricing and tax avoidance through the indirect transfer of assets – had not been addressed in the BEPS package.[26] Fourteen developing countries were subsequently invited to participate in the OECD’s Committee on Fiscal Affairs and a series of regional consultations were undertaken. However, this occurred late in the process, when the list of priority actions on BEPS had already been agreed. We were only able to find evidence of two measures introduced into the BEPS package as a direct response to developing country concerns.[27]

Similarly, the standards on automatic exchange of information were developed without consulting developing countries. Commentators have pointed out that they are both technically demanding and resource-intensive for developing countries to implement. The G20 Development Working Group recognises that implementing the automated system will be challenging for some tax administrations and may need to be a long-term objective.[28]

To help developing countries adopt the new standards, the Global Forum has recommended piloting automatic exchange of information in selected countries; DFID, with support from the Treasury and HMRC, is leading the pilot in Ghana. With funding from DFID and other donors, the OECD is also working with other international organisations and regional tax bodies to develop guidance (‘toolkits’) on areas of interest to developing countries which were not included in the final package of BEPS measures.[29]

In all, DFID’s objective of influencing the G20 tax standards to reflect the needs of developing countries was not achieved. The BEPS measures do not address a number of issues of concern to developing countries, and the automatic exchange of information standards remain largely beyond the capacity of many tax authorities to implement.

DFID has not supported the wider international tax reform agenda

Despite recent efforts to broaden participation, the G20 and the OECD are not well positioned to represent the needs and priorities of developing countries. Most of the stakeholders we interviewed – including NGOs, regional organisations, multilateral agencies and developing country governments – were in agreement that the OECD and G20 countries still largely set the international tax reform agenda.

Many developing countries and development NGOs have therefore called for more fundamental reforms to the structure of the international tax system, to make it fairer for developing countries. They have also called for changes to the international architecture where these issues are debated, to make it more representative. DFID has not so far supported this wider reform agenda.

At the 2015 UN Financing for Development conference in Addis Ababa, the G77 (which represents 126 countries) called for the UN Tax Committee to be upgraded from its current status as a ‘group of experts’ to an intergovernmental body with balanced geographical representation. Their objective was to create a new, more legitimate counterpart to the OECD, as a forum for debating more fundamental tax reforms. The proposal was rejected by the UK and other OECD countries, who argued that it would be duplicative and ineffective.[30]

DFID informed us that it has chosen to work with the G20 and the OECD in order to make use of their technical capacity and the political momentum behind the existing international tax processes. While the UK is a long-standing member on the UN Tax Committee and fronts an HMRC representative at meetings, it does not believe that the committee would be an effective standard-setting forum. We accept that this is a reasonable assessment of the current situation, and that DFID is bound by UK government policy positions.

In all, attempts to make the G20 tax reform agenda truly inclusive have not been successful; a number of other donors we spoke to recognised the need for a more representative international tax architecture and several were providing funding to the UN Tax Committee to help it become a more effective body.

DFID’s push for more capacity building support through the G8/G20 had limited success

One of DFID influencing objectives for the 2013 G8 summit was to secure more international funding for capacity building on tax. DFID achieved the inclusion of references to capacity building in the G8 communiqué, including support on international tax and a long-term commitment to sharing tax expertise, together with a paragraph on “tax and development”.[31]

Our review of internal correspondence between DFID and the Treasury in the lead up to the summit, reinforced by interviews with officials, confirms that DFID helped to achieved these outcomes, working through the UK Presidency.

However, the commitments were limited in their ambitions. In the communiqué, G8 countries agreed to continue, rather than increase, their capacity building support. This fell short of DFID’s objective. As a result, DFID remains today the largest donor for both the OECD Tax and Development Programme and the Global Forum technical assistance programme.[32]

DFID’s support to the Addis Tax Initiative shows a top-down approach to developing country participation

In the face of reluctance from some donors to make specific funding commitments on international tax, DFID switched its efforts to other channels. Working with like-minded donors, it supported the Addis Tax Initiative, which was launched at the UN Financing for Development Conference in July 2015, to encourage the mobilisation and effective use of domestic revenues for national development.[33]

DFID initiated early discussions on the Initiative in various forums, including the OECD Tax and Development Taskforce and the G20 Development Working Group. It succeeded in securing a reference to international tax, but its attempts to negotiate a commitment to doubling donor support for domestic resource mobilisation ended in compromise, with donor signatories committing to ‘collectively’ doubling their assistance, allowing greater flexibility on individual contributions.

While the Addis Tax Initiative is described as a partnership, the preparations leading to its launch were donor-led and top-down in nature, rather than consultative or based on an analysis of developing country needs. Progress in getting DFID partner countries to sign up to the Initiative has been slow. To date, only nine DFID priority countries have done so, even though our survey suggests that there is appetite from developing countries for more donor support for domestic resource mobilisation.

Findings: Has DFID helped developing countries implement international tax standards?

DFID’s capacity building support is helping partner countries adopt international tax standards and practices 

DFID supports implementation of the international tax standards and practices through two centrally managed programmes: i) the HMRC Tax Expert Unit and ii) the Tax Transparency Programme, which are starting to deliver some early results.

The Tax Transparency Programme funds the OECD and the World Bank to support developing countries in adopting and implementing international tax standards and practices, both on transfer pricing and exchange of information. The OECD has also reviewed tax incentive regimes in selected countries.[35] While the programme is still at an early stage, there have been a number of achievements up to mid-2016:

• 11 new countries have joined the Global Forum, which oversees the exchange of information standards.

• 14 countries have adopted national regulations on exchange of information and, in doing so, have passed Phase 1 of the Global Forum peer review process.[36]

• Uganda, Kenya, Jamaica and Senegal have joined the Convention on Mutual Administrative Assistance in Tax Matters, while other countries have signed bilateral exchange of information agreements.

• 14 countries have begun implementing action plans for legislative changes and administrative improvements for transfer pricing and related controls. Ethiopia adopted transfer pricing legislation in October 2015.

• Five countries (including one DFID priority country, Ghana) and one region (the Southern African Development Community) have received reviews of their tax incentives for investment by the OECD.

• In some cases, new specialist units have been established within national tax administrations.[37]

It remains unclear whether low-capacity countries will be able to implement the standards effectively

Despite these early results, some of the partner country stakeholders we interviewed expressed doubts that technical assistance in these highly specialised areas was likely to be effective in low-capacity environments. Aspects of the new standards are technically demanding and resource intensive to implement, and it is not yet clear whether DFID’s partner countries will be able to apply them effectively.

Among countries that have passed the necessary legislation on exchange of information, only a small number have begun requesting information from international partners. In Ghana, which has received capacity building support from the Global Forum since 2011, the tax revenue authority informed us that its staff remain unsure of when it is appropriate to make an information request. With automatic exchange, tax administrations in least developed countries may lack sufficient in-house capacity to make use of the high volumes of information that they would obtain.

Sustainability is also an issue. In the area of transfer pricing, nearly 200 individual cases were resolved in four countries (Kenya, Colombia, Vietnam and Zambia) in the first two years of the Tax Transparency Programme, leading to the recovery of approximately £85 million in revenue. The success of these cases shows the benefits of having the necessary legal framework in place. However, these cases were resolved with direct assistance from international tax experts. Even in countries that have received assistance for several years, there is little evidence that the tax authorities have the in-house capacity to carry out such audits without external support.

In our interviews, a small number of donor and developing country representatives criticised the OECD for having a narrow and overly technical approach to capacity building, without taking sufficient account of the underlying capacity constraints.

The new Tax Administration Diagnostic Assessment Tool may help with this sequencing challenge. Launched in late 2015, this tool offers a method of assessing the capacity of national tax administrations against a common set of indicators. DFID has contributed £2 million towards its development and for assessments in DFID partner countries. Such standardised diagnostic tools already exist in other areas of public financial management. They can help to build a common understanding among donors and partner countries on capacity building needs and priorities.

The benefits to developing countries of adopting international standards may have been oversold

The bulk of DFID’s capacity building support through the OECD has gone towards promoting the implementation of the international tax standards and principles in developing countries. This support was designed to be demand-led. The majority of tax authority stakeholders we interviewed from DFID’s partner countries confirmed that they need capacity building support to implement them.

Yet in practice, demand from DFID’s priority countries for OECD assistance has been mixed. According to programme documentation, DFID partner countries have taken up the offer of support on transfer pricing. However requests for assistance on exchange of information have instead come mainly from more advanced countries, such as Jamaica and Colombia, which are not priority countries for DFID.

In their monitoring reports, DFID and the OECD concluded that this reflected a lack of awareness of the benefits of the new standards in fighting tax avoidance and evasion on the part of developing countries. To address this, DFID provided the OECD with a further £0.5 million to launch the ‘Africa Initiative’, to promote the benefits of information exchange and build political support from African leaders for its implementation. A similar process of awareness-raising was also used for BEPS.[38]

Some awareness-raising among developing countries may have been justified in order to identify latent demand for the new standards. Yet a wide range of international and developing country stakeholders expressed concern to us that the benefits to developing countries of adopting international standards may have been oversold.

While hailed by G20 countries and the OECD as a major success, some UK-based NGOs and privatesector representatives expressed concerns to us that the BEPS measures will not have the expected results. These concerns are echoed in the literature. Many of the BEPS actions are non-binding. They focus only on tax avoidance, leaving most tax planning structures intact, and still permit aggressive tax competition to take place between jurisdictions.

Exchange of information between national authorities may not be enough to prevent wealthy individuals from developing countries using tax havens for tax avoidance and evasion. In our interviews with DFID officials and in internal documentation, DFID also recognised that in countries where the space to scrutinise public officials is small, making beneficiary ownership information public is important to support social accountability. Observers also point to the problem of corruption in national tax administrations and the lack of effective sanctions or asset recovery mechanisms as likely to undermine the value of the new international tax standards to developing countries.

Many developing countries also grant multinational corporations overgenerous tax incentives that are harmful to their economies. One report even goes so far as refer to tax incentives as “tax evasion with an official stamp”.[39] If multinational corporations have little tax liability in the first place, then implementing the BEPS measures is unlikely to raise significant additional revenue.

Implementing new tax standards could draw resources away from more important priorities

While studies suggest that international tax avoidance and evasion have a substantial impact on developing countries, the revenue gains from addressing domestic tax issues are likely to be much higher. Our survey and our case studies confirm that many developing countries consider strengthening their domestic tax systems as a higher priority.

There is therefore a risk that implementing new tax standards could draw limited national capacity away from more important domestic tax priorities. Most of DFID’s spending on tax and development goes towards reforming national tax systems. However, DFID and its implementers appear to have given little consideration to the question of how to sequence the introduction of international tax standards with more basic reforms of national tax systems.

The G20 Development Working Group recognises that capacity building on international tax needs to be better coordinated, sequenced with more basic reforms, and better matched to the individual needs and conditions of countries, if it is to achieve sustainable results.[41]

Findings: Has DFID promoted cross-government coordination on international tax?

DFID has worked well with the Treasury and HMRC to influence the G7/G8 and G20 agenda

DFID worked closely with Treasury (as lead department) and HMRC to shape UK positions ahead of the G8 summit in 2013 and around the G20 tax agenda. The Treasury drew on a range of DFID inputs, including a concept note on the G8 and developing countries, in its tax and development work. We saw evidence that strong cross-government working had contributed to effective UK advocacy at the G8 summit in 2013.

The relationship between the three departments is now well established, with a mixture of formal and informal interactions between the responsible teams.[43] DFID is regularly consulted by Treasury on discussions in the G20 Finance Track. Recently, the two departments worked together on a UK proposal on tax and development, which helped to secure support of the principles of the Addis Tax Initiative in the G20 Finance Ministers’ communiqué.[44]

The HMRC Capacity Building Unit is potentially a good model for collaboration on technical assistance at the country level

The International Development Committee has highlighted DFID’s good work in the tax area and recommended that more should be done, including in partnership with HMRC.[45] To that end, in 2014 DFID provided funding to enable HMRC to establish a Capacity Building Unit. HMRC now finances the Unit from its own aid budget, with estimated funding of £22.9 million over ten years to 2024. The Unit provides technical assistance on domestic tax. It complements a three-person HMRC Tax Expert Unit, which works on international tax issues.

Through the Capacity Building Unit, technical experts from HMRC are deployed to DFID priority countries in both short- and long-term advisory roles. These deployments are made in response to requests from DFID country offices, to complement DFID tax programmes. After a slow start, longterm advisors are now in place in Ghana, Tanzania and Ethiopia, where they are fully integrated into DFID bilateral programmes and able to call on additional specialised inputs from HMRC through short-term missions.[46] Long-term advisors also support Pakistan (working from both London and Pakistan) and the African Tax Administration Forum. Other countries have received support solely through short-term missions.[47]

Our country case studies show that this form of collaboration has the potential to work well. Embedding HMRC advisors in DFID programmes ensures that HMRC’s technical expertise is used within a long-term capacity development strategy aligned to each partner country’s needs and priorities.

While it is too early to assess results, tax revenue authorities in Tanzania, Ghana, Ethiopia and Pakistan have expressed their appreciation for the opportunity to work with UK peers and have welcomed the combination of long- and short-term assistance.[48] In Ethiopia, where HMRC has been engaged since 2006, support to the Revenue and Customs Authority has reportedly performed well.

However, there have also been challenges with this partnership. HMRC has sometimes had difficulty finding appropriately qualified staff to respond to partner country needs. Scoping and establishing relationships have taken longer than expected. Adapting to DFID’s decentralised structure has also been challenging for HMRC. When the Capacity Building Unit was launched, DFID bilateral tax programmes were at different stages of development and varied in their appetite for its support. These limits on both supply and demand have meant that progress has been slow. Our survey also indicated that both national tax authorities and DFID advisors in partner countries were less convinced that one-off short missions from HMRC experts add much value, in the absence of a wider engagement.

Centrally, DFID is beginning to provide more active support to other government departments on tax and development

During our interviews, other UK government departments indicated that they valued the knowledge on country context available in DFID country offices. Yet we found limited evidence that other departments looked to DFID centrally for its expertise on complex programming, governance and capacity building, or that DFID had attempted to cultivate this role.

In recent months, the DFID tax team has recognised the need to strengthen its support to other government departments and country offices. Under the Chinese G20 Presidency, development issues are being mainstreamed into the Finance Track, which means that the tax and development work is being led by the Treasury (DFID maintains the domestic resource mobilisation co-facilitator role in the Development Working Group). In response, the DFID tax team has stepped up its advisory role, making a significant contribution to the Treasury’s tax and development submission to the G20 in 2016.

With the HMRC Capacity Building Unit, DFID’s tax team initially focused on facilitating partnerships with DFID country programmes. Once these had been established, the expectation was that DFID’s central team would provide guidance on strategic issues, impact monitoring and pipeline management.[49] DFID was slow to take up this role. In 2015, Treasury initiated a strategic framework for capacity building for the HMRC Capacity Building Unit, while HMRC decided to use its own monitoring and evaluation system.[50] DFID has now recognised that it has an important contribution to make on those issues, and has stepped up the level of its support to Treasury and HMRC.

Findings: Has DFID promoted UK policy coherence on international tax?

DFID does not have a considered approach to policy coherence for development in tax

Alongside other donors, the UK has committed itself to pursuing ‘policy coherence for development’ by promoting greater coherence across its public policies, to increase the opportunities for developing countries[51]. In its 2015 Beyond Aid report,[52] the International Development Committee recommended that DFID make policy coherence for development a higher priority.

The Addis Tax Initiative also contains a broad commitment on policy coherence, although this was the result of advocacy by other donors. It states: “All participants will ensure that relevant domestic tax policies reflect the joint objective of supporting improvements in domestic resource mobilisation in partner countries and applying principles of transparency, efficiency, effectiveness and fairness.”[53]

Based on our interviews with DFID and other government departments, consultation with civil society organisations and a review of internal documentation, we find that DFID has not actively pursued policy coherence for development in the international tax arena. DFID and Treasury officials make the argument that, in light of the UK’s strong commitment to tax transparency, UK domestic policy and the international development agenda are already aligned.

However, the literature suggests that there are areas of tension between UK policies and developing county interests in the broader tax arena, around issues such as international tax competition and bilateral tax treaties. In our review of internal documents and email correspondence, we found no evidence that DFID either raised or considered raising these issues in its discussions with other departments. As the UK government representative on the G20 Development Working Group, DFID rejected calls from the IMF and the OECD to assess the ‘spillover impacts’ of UK tax policies on developing countries , reflecting the Treasury’s view that this would be impractical and
time-consuming.

We take no view on the merits of UK policy positions on ‘spillover analysis’ nor of international tax more generally, which are beyond our mandate. We also recognise that, once a UK policy decision is taken, it is binding on DFID. However, given that tax is highlighted in the Aid Strategy as a ‘beyond aid’ priority area, we would have expected to see evidence of DFID taking active steps to understand how UK tax policies impact on developing countries, and on this basis, making informed judgments about whether to discuss possible areas of tension in cross-government dialogue. In the absence of any evidence of a structured decision-making process of this sort, we conclude that DFID is not actively pursuing policy
coherence in the area of international tax.

This finding echoes the conclusions of the OECD’s 2014 peer review, which found that the UK government lacked “a comprehensive approach to ensuring its development efforts are not undermined by other government policies”.[57] Some other donors have been bolder in adopting a more coherent approach to international tax.[58]

Findings: Has DFID generated and applied evidence and learning?

Both the international tax reform agenda and working cross-government to influence international standards are relatively new areas for DFID. We assess the extent to which it has drawn on evidence and learning to improve its work.

DFID has invested in research, but data gaps remain a significant obstacle to effective international tax policy-making

DFID recognised at an early stage that a lack of solid evidence on the significance of international tax issues for developing countries was an obstacle to effective policy-making. Poor quality data had contributed to a divergence of opinion on how to proceed.

In 2009, DFID commissioned the Oxford University Centre for Business Taxation to assess the quality of evidence on the scale and effects of tax avoidance and evasion in developing countries.[59] Its report concluded that most existing estimates were not based on reliable methods and data.[60] Our literature review indicates that little progress has been made on filling the data gaps since then.

DFID supported the launch of the International Centre for Tax and Development (ICTD) at the University of Sussex with a £3.5 million contribution (2010-16). Feedback from donor representatives and academics indicates that ICTD’s research is well regarded. Alongside other work, DFID had intended that ICTD would generate new evidence on international tax. However, the absence of reliable data from low- and middle-income countries has limited ICTD’s ability to generate new findings. Donor, academic and developing country representatives agree that limited progress can be made until more primary data is generated from developing countries. ICTD has begun addressing this gap by creating a new cross-country tax dataset and experimenting with collaboration between researchers and tax authorities.

DFID has not used research to push the boundaries of debate internationally or internally

An internal document from 2009 notes DFID’s intention to invest in research as a way of bringing stakeholders together around a common understanding of tax and development. While DFID has invested in some research and evidence collection, our review of internal documentation shows that it has relied mainly on diagnostic work produced by the OECD and other international organisations to inform its influencing approach. These diagnoses have focused primarily on the G20 tax reform agenda.

Within the growing body of international tax literature produced by international organisations, NGOs and academics, there is lively debate on the adequacy of the G20 tax initiatives and the potential for alternative models of international taxation.[61] In our documentary analysis we found little evidence that DFID had actively engaged with these debates in setting its own priorities or in dialogue with other UK government departments or external partners.

DFID has not drawn sufficiently on its in-country knowledge to inform its approach to international tax

One of DFID’s areas of comparative advantage is its strong in-country network. Our survey revealed a good level of interaction between DFID country offices and national tax authorities. Yet we saw little evidence of DFID’s central team drawing on its country contacts and knowledge to inform the design and delivery of its international tax influencing activities and central programmes. Only three of the 18 country offices who responded to our survey had worked with the central policy teams on G20 issues. These were Kenya and Ghana – the two pilot countries under the Global Forum – and South Africa, a G20 country that already participates in the international tax standard-setting processes.

DFID’s Financial Accountability and Anti-Corruption Team argued that, at that stage, country offices had little knowledge of the G20 agenda and therefore could not have inputted into the UK position. However, we found no evidence of DFID drawing on knowledge from its bilateral programmes to assess whether the proposed G20 solutions would be relevant to its partner countries’ needs and priorities. This lack of consultation may have also contributed to DFID’s overestimating the demand for capacity building support from its priority countries.

Lack of monitoring of influencing work creates a risk of over-estimating achievements

DFID did not set itself an explicit set of influencing objectives for its international tax work. Nor did it establish any method of assessing its impact. In the absence of clear objectives or monitoring arrangements, we have observed in some DFID interviews and internal documents a tendency for DFID to be over-optimistic in the results it claims to have achieved through its efforts.[62]

Engaging with international standard-setting is an important part of the ‘beyond aid’ agenda. The formulation of objectives and the measurement of results need to be approached with the same level of rigour that DFID would apply to a more traditional aid programme. We note that since the start of our review, the DFID team has begun developing an evidence-based position paper and a theory of change on how they intend to use financial and non-financial instruments to meet UK government commitments on tax and development by 2020.

Findings: Has DFID’s approach to international tax been good value for money?

DFID has made good use of limited advisory resources and helped build the capacity of its international partners

In the international tax area, DFID has made efficient use of a small investment of resources (£38.9 million committed over fourteen years). With inputs from just three advisors, it was able to participate actively in a wide range of international forums and processes, albeit in pursuit of relatively narrow objectives.

DFID has also combined its spending and influencing activities efficiently. At the 2013 G8 summit in Lough Erne, one of DFID’s main objectives was to secure international commitments for increased capacity building for developing countries. It backed this with funding commitments of its own: the Tax Transparency Programme and the HMRC Capacity Building Unit were launched in parallel to the G8/G20 commitments. DFID maximised its reach by working through others (the OECD, the World Bank and the IMF). It also made effective use of programme resources to support its influencing work.

Good collaboration with the Treasury and HMRC has also proved cost-effective. Building up HMRC’s Capacity Building Unit is consistent with the goal in the UK Aid Strategy of making more use of complementary skills across government.[63]

DFID has used its funding to build up the capacities of its international partners to help deliver its goals. It was willing to take the risk of being one of the first donors to support the OECD’s tax and development work and the African Tax Administration Forum. It took advantage of the OECD’s resources and expertise, helping to establish a presence in developing countries and build up diagnostics on capacity building. This in turn informed the development of the G20 tax agenda.

Similarly, DFID has invested in building HMRC’s capacity to operate in developing countries. This has enabled HMRC to scale up its activities and adopt a more structured and strategic approach to technical assistance.

HMRC’s early use of funds for capacity building has raised value for money concerns

As a relatively new player in development assistance, HMRC’s early use of funds did not offer good value for money. Given funds estimated at £22.9 million over ten years, its scaling up plan proved overambitious. In its first year of operations, half of the funding provided by DFID (£1.17 million) was spent on training 15 new HMRC tax experts for domestic roles, in order to release existing staff for deployment abroad.[64] As it transpired, HMRC has never achieved more than half of its planned deployment and is unlikely to be able to deploy as many as 15 full-time equivalent advisors in the near future.

This highlights a key value for money risk as departments take on new aid delivery roles. It will take time to establish the systems, capacities and programmes needed in order to spend development assistance effectively in new country contexts. HMRC acknowledges that it had limited experience of delivering programmes across a range of developing countries. Having relevant technical expertise may not be sufficient to achieve good value for money, unless departments also draw on DFID’s experience with effective aid programming.

Short-term and peer-to-peer technical assistance must demonstrate better value for money

The other main value-for-money concern relates to short-term technical assistance – that is, one-off short missions by technical experts to provide advice or training, which are offered by both HMRC and the OECD. In our survey and our country case studies, officials from national tax authorities expressed doubts that such missions add much value – a view confirmed by some DFID in-country advisors. In many countries, revenue authorities do not have the capacity to absorb highly specialised, short-term technical inputs or one-off training sessions.

This is an important consideration as the UK government increases its use of in-house expertise for development assistance. Counterparts appreciate the opportunity to interact with their UK peers. However, there is still much to be learnt about how to use peer-to-peer assistance to best effect – a point acknowledged by both DFID and HMRC. This will be particularly important for the HMRC Tax Expert Unit, which provides most of its specialist expertise in the form of short-term missions.

Recommendations

We recognise the importance of the UK aid programme engaging ‘beyond aid’, by influencing international norms and standards. We welcome the contribution that DFID has made to date on international tax. The following recommendations are intended to help DFID strengthen its future engagement on this important issue.

Recommendation 1: Learning on international tax issues

DFID should make better use of its in-country work on both tax and anti-corruption to inform its influencing efforts and prioritise its programming on international tax.

Problem statements:

• Fighting tax avoidance and tax evasion in developing countries requires both tax and anticorruption expertise, with solutions likely to extend beyond the adoption of international standards.

• DFID has not made effective use of evidence from its bilateral country programmes to analyse which international tax reforms are most likely to deliver the greatest benefits to developing countries, given underlying capacity and political economy conditions.

• The question of how to prioritise and sequence international tax initiatives with efforts to strengthen national tax systems has not been properly addressed.

Recommendation 2: Cross-government working

DFID should be more proactive in ensuring that other departments engaging in capacity building on tax and more generally, are able to draw on its experience of effective capacity building approaches and its knowledge of country contexts. This will require closer collaboration between departments, both at headquarters and in country.

Problem statements:

• DFID’s tax team has been slow to articulate how it can support other departments’ work on tax and development.

• The Aid Strategy notes the importance of using the complementary skills available across the UK government in the delivery of aid.

• Departments that are scaling up their aid expenditure need to acquire the capacity to design and deliver sound programmes and monitor and evaluate their results.

• While other government departments are accountable for the use of their own aid budgets for tax, they are likely to achieve better results if they draw on DFID’s experience of effective capacity building and of the context in its partner countries.

Recommendation 3: A strategic approach to influencing

DFID should adopt a more systematic approach to influencing and cross-government working around international tax, with a stronger strategy and explicit objectives that are adequately resourced and properly monitored.

Problem statements:

• DFID’s opportunistic approach to influencing on international tax has limited its impact.

• If DFID is to be effective in its ‘beyond aid’ influencing work, it needs explicit strategies and objectives and an appropriate level of investment in monitoring results.

Recommendation 4: Policy coherence for development

DFID should take a more active approach to promoting policy coherence for development on international tax by assessing the impact of UK tax policies and practices on developing countries and deciding whether to raise any potential points of tension in cross-government dialogue.

Problem statements:

• As a signatory to the Addis Tax Initiative and the Busan Partnership for Effective Development Cooperation, DFID has made a commitment to promoting policy coherence for development, but it has yet to develop an explicit approach to following through on this commitment.

• The UK government does not yet have a full view of where it tax policies might have negative ‘spillover’ effects on developing countries.

Learning Frontiers

Under the Aid Strategy, the UK government has committed to working together in a number of ‘beyond aid’ priority areas, in addition to international tax. The objective is to influence international norms and standards, to promote more opportunities for developing countries.

As a learning review, we are interested in identifying areas where more learning is needed, in order to inform continuing improvement in the UK aid programme. We call these ‘learning frontiers’. Here, we offer a number of learning frontiers that we believe would be useful both to DFID and other departments in their ‘beyond aid’ engagements.

Cross-government working

• Producing a synthesis of good practice on effective capacity building approaches in low-capacity environments.

• Gathering evidence on how to make peer-to-peer assistance more effective and good value for money.

Policy coherence for development

• Research to identify areas where UK policies and practices on tax, trade and investment could be adjusted to the benefit of developing countries without substantial compromise to UK interests.

• Clarifying when and how DFID’s central policy teams should raise with other departments the impact of their policies on developing countries.

Annex 1 Detail of Scoring

Question 1 Relevance

How coherent and relevant is DFID’s approach to the global issues of cross-border tax avoidance and tax evasion?

DFID’s work on international tax has not been based on close consultation with its partner countries or analysis of their needs.

DFID actively promoted the benefits of implementing the new international tax standards to its partner countries through its capacity building work. Stakeholders are concerned that these benefits may have been oversold. It is unclear whether the standards can be implemented effectively or would generate much additional revenue, given more basic problems with national tax systems.

DFID lacks a considered approach to promoting policy coherence for development on tax. It has not attempted to identify tensions between UK tax policies and developing country interests, nor made explicit decisions about what issues to raise in dialogue with other departments.

We have therefore given DFID’s international tax work an amber-red rating for relevance.

Question 2 Effectiveness

How effectively has DFID contributed to addressing the global issues of cross-border tax avoidance and tax evasion in a way that benefits developing countries?

DFID has taken advantage of the opportunities presented by the G20 and OECD tax initiatives, using a range of influencing channels, and effectively combined influencing with aid programming.

It helped developing countries participate in various international tax processes; however, most stakeholders agree that they gained limited practical influence over the new standards.

DFID’s efforts to secure more donor funding for international tax have been partially successfully. Through the Addis Tax Initiative, it helped secure a commitment from a group of donors to collectively doubling their funding for tax and development.

DFID’s capacity building support has achieved some early results: various partner countries have been assisted to adopt new regulations, conclude international agreements, develop action plans or establish specialised tax units. However, there remains some doubt as to the impact and sustainability of these results, given capacity constraints in national tax systems.

DFID’s partnership with HMRC is potentially a strong model for cross-government collaboration on capacity building. It enables DFID country programmes to call on specialist expertise from HMRC, while ensuring that HMRC’s technical assistance is anchored in a broader strategy aligned to each country’s needs and priorities.

However, we have concerns about the value for money of some of HMRC’s early assistance and of its short-term technical assistance.

We have given DFID’s influencing efforts a green-amber score for effectiveness, in recognition of good cross-government collaboration and some positive results on capacity building.

Question 3 Use of evidence and learning

To what extent is DFID generating and applying evidence and learning to support its approach to addressing the global issues of cross-border tax avoidance and tax evasion?

DFID is aware of the evidence gaps on international tax and has made some worthwhile investments in research. However, it has not drawn on available evidence and learning to inform its activities or to push the boundaries of policy dialogue on international tax. Nor has it drawn sufficiently on the knowledge available in its country offices to inform its central approach to international tax. It has also not sufficiently cultivated its advisory role across the UK government, either on tax and development or on capacity building.

A lack of clear objectives, explicit strategy and monitoring arrangements for its influencing have limited DFID’s ability to learn lessons and have led to a tendency to be over-optimistic in the results it claims to have achieved. We have therefore given DFID an amber-red score for learning.

Annex 2 Glossary

Base Erosion and Profit Shifting

The OECD has developed the term Base Erosion and Profit Shifting (BEPS) for “tax planning strategies that exploit … gaps and mismatches in tax rules to artificially shift profits to low or no-tax locations where there is little or no economic activity, resulting in little or no overall corporate tax being paid.” The OECD observes that “BEPS is of major significance for developing countries due to their heavy reliance on corporate income tax, particularly from multinational enterprises (MNEs).”[65] In practice, the term is synonymous with the OECD’s project to reform international tax rules, which ran from 2013 to 2015, and is now in its implementation phase.

Indirect transfers of assets

The IMF explains the issue of indirect transfer of assets as follows: “At issue is the possibility that the owners of some asset with respect to which a capital gain arises can avoid tax in the country to which that asset—perhaps a telecom or mineral license—inherently relates, by holding it through a chain of companies and then selling the claim in a low tax jurisdiction”. Such transactions can involve the host country receiving little or no revenue when substantial gains are realized on assets located there.[66]

Developing countries’ concerns are visible through the court cases pursued by countries such as Uganda and India, and also through the introduction of new legislation in countries such as Mozambique and Kenya.[67] Court cases are typically concerned with the indirect sale of assets / licences in the natural resource and telecommunication sectors, with the transaction taking place offshore. Domestic law reforms to capital gains tax may focus explicitly on the extractive sector.[68]In addition to the domestic law reforms that can help tackle this form of tax avoidance, international tax rules are relevant in three ways:

i. Most bilateral tax treaties leave developing countries vulnerable to this kind of tax avoidance, because they do not include the UN and OECD model treaty provisions designed to prevent it.

ii. Developing countries need information because they may not know that an indirect transfer is happening if it takes place abroad.

iii. Mutual assistance between tax authorities is needed because once the taxpayer has sold its assets and exited the developing country, the tax authority has no power to compel it to pay taxes.

Spillover analysis

International organisations have recommended that developed countries undertake ‘spillover analysis’ when making changes to their tax systems with the potential to have an impact on other countries, especially developing countries.[69]In 2011, the IMF, OECD, UN and World Bank suggested that such analyses “should be published for the international community to reflect upon – at a minimum, to enable developing countries to respond with parallel changes to their own systems if that would be helpful in protecting their revenue bases. Ideally, a ‘baseline analysis’ along these lines would be undertaken immediately.”[70]

One example highlighted by the IMF is reforms to the taxation of multinational companies in the countries they are headquartered, such as those implemented by the UK in recent years, which exempt them from paying tax on their overseas earnings. By allowing businesses to retain the benefits of any tax advantages granted by host countries, these reforms are likely to intensify tax competition between developing countries.[71] Ireland and the Netherlands have both commissioned ‘spillover analyses’ of the impact of their tax treaty networks on developing countries.[72]

Tax avoidance

Tax avoidance and tax planning are both terms that refer to using a country’s tax regime, or tax regimes in different countries, to minimise the tax payable. Views differ as to what constitutes ‘legitimate’ tax planning and ‘abusive’ tax avoidance, and the OECD, whose remit focuses on international tax, describes tax avoidance as “hard to define.”[73] A common formulation of the difference between the two is that tax avoidance complies with the strict letter, but not the spirit, of the law, while tax planning complies with both. According to HMRC, “tax avoidance involves bending the rules of the tax system to gain a tax advantage that the Government or Parliament never intended. It often involves contrived, artificial transactions that serve little or no purpose other than to produce this advantage.”[74]

Practices by multinationals that attract these labels often involve shifting pre-tax profits into low-tax jurisdictions, including non-cooperative jurisdiction and tax havens. Two such practices are tax treaty shopping and transfer pricing manipulation. HMRC explains treaty shopping as “Taking advantage of the network of double taxation agreements [between countries] to obtain a more advantageous position than the facts warrant.”[75] Transfer pricing, the system through which multinational firms’ taxable profits are allocated between the different countries in which they operate, can be manipulated to shift firms’ pre-tax profits into low-tax countries, reducing their overall tax bills.

Tax evasion

Tax evasion is described by the OECD as “a term that is difficult to define but which is generally used to mean illegal arrangements where liability to tax is hidden or ignored, ie the taxpayer pays less tax than he is legally obligated to pay by hiding income or information from the tax authorities.”[76]Tax evaders may hold assets overseas, particularly in less transparent jurisdictions, in order to conceal them from tax authorities. Tax information exchange is therefore the primary tool for combatting international tax evasion.

Though multinational firms may commit tax evasion, for example through transfer mispricing (when a cross-border tax arrangements break applicable norms internationally or at domestic law)[77] it is domestic firms and individuals that are the focus of efforts to combat tax evasion in developing countries.

Tax havens and uncooperative jurisdictions

While it is widely accepted that some jurisdictions make aspects of their tax system available for people and companies to avoid and evade their tax liabilities in other countries, efforts to arrive at a more concrete definition of ‘tax havens’ have always been controversial. In 1998, the OECD set out a number of factors for identifying tax havens. The four most important were: (i) no or nominal tax on the relevant income; (ii) a lack of effective exchange of information; (iii) a lack of transparency; and (iv) no requirement that a person must have substantial activities in a jurisdiction to claim tax residence there.

In order to avoid being listed as an uncooperative tax haven, jurisdictions that met the first three criteria were asked by the OECD to make commitments to implement the principles of transparency and exchange of information for tax purposes,[78] however the organisation struggled to maintain the political consensus needed to compel jurisdictions to reform completely.[79]

In 2009, the OECD stopped listing jurisdictions as uncooperative, and the process was overhauled with G20 support. The Global Forum was established as a satellite body to the OECD to peer review jurisdictions compliance with information exchange standards, rating jurisdictions as partially, largely or fully compliant. Some countries maintain their own blacklists, with varying levels of objectivity, and in 2015, following the Luxembourg leaks, the European Commission published a ‘black list’ compiled from those of its individual member states.[80] The UK has resisted EU calls to reintroduce a similar list, following the Panama papers.[81]

While the term ‘tax haven’ is still widely used in common parlance, it is rarely used in policy processes, which focus on jurisdictions’ secrecy characteristics, as opposed to tax rates. The Tax Justice Network prefers the term ‘secrecy jurisdiction’, and publishes a bi-annual Financial Secrecy Index to help identify jurisdictions in need of reform.[82]

Tax incentives

Despite a chorus of opposition from international organisations, NGOs and academics, developing countries have introduced a growing array of tax exemptions or reductions for investors, such as short-term ‘tax holidays’ and tax-exempt special economic zones.83 Estimates of the cost of tax exemptions, by governments, international organisations and nongovernmental organisations, run to several percent of GDP.[84]

For example, in a series of workshops held by the African Tax Administration Forum with tax revenue authorities, participants stated that “there is a proliferation of tax incentives and exemptions across the regions and they are often not evaluated in terms of cost-benefit analysis… tax incentives and exemptions have been over-emphasized as investment promotion tools.”[85]

A report prepared for the G20 in 2011 by the IMF, OECD, United Nations, and World Bank – organisations that have in the past encouraged countries to adopt tax incentives – concluded that “where governance is poor, they may do little to attract investment – and when they do attract foreign direct investment (FDI), this may well be at the expense of domestic investment or FDI into some other country. Tax-driven investment may also prove transitory.”[86] Through the Tax and Development Task Force, the OECD has produced a set of principles to enhance the transparency and governance of tax incentives for developing countries. It has reviewed a number of developing countries against these principles.[87]

Transfer pricing comparables

Transfer pricing, the system through which multinational companies’ pre-tax profits are allocated between the countries in which they operate, focuses on the transactions that take place between different parts of a multinational company around the world. The OECD’s approach to transfer pricing relies on finding comparable transactions that take place in the free market as a reference point. According to the UN Tax Committee, “Proper comparability is often difficult to achieve in practice…It is often extremely difficult in practice, especially in some developing countries, to obtain adequate information.”[88] This is partly because the commercial databases used for transfer pricing analysis have poor coverage in developing countries, and also partly for structural reasons: in many instances, there are no comparable transactions in developing countries that can be used to apply transfer pricing rules.

The 2013 Lough Erne G8 declaration made a specific reference to this point, asking the OECD to “find ways to address the concerns expressed by developing countries on the quality and availability of the information on comparable transactions that is needed to administer transfer pricing effectively.”[89] This work is ongoing, but critics argue that this problem calls for a fundamental rethink, using approaches that do not rely on comparables, or even abandoning transfer pricing altogether.[90]

Withholding taxes

In many countries, employers are required to withhold their employees’ income tax from their salary payments. Similarly, in many countries a tax must be withheld on payments of dividends, interest payments, royalty fees and technical fees, especially when the recipient of these payments is overseas. Withholding taxes serve two purposes: they give the country from which the fee is paid (often a developing country) some opportunity to tax income earned in their borders by foreign residents, and they discourage multinational firms from extracting income and shifting profit using these types of transactions.[91]

Withholding tax rates have been declining over time, both in countries’ domestic law and in bilateral tax treaties, which impose limits on the rates that countries can enact. OECD policy, as embodied in its model bilateral tax treaty, is to allow limited withholding taxes on dividends and interest payments. The OECD model does not allow source countries to tax royalties or service fee payments at all, while the UN equivalent permits withholding taxes on royalties and will soon include a clause on technical fee payments as well. Some commentators have argued that developing countries should retain higher withholding tax rates.[92] As part of the BEPS project, the OECD considered the use of withholding tax on interest payments to discourage tax avoidance through debt financing, but instead arrived at a statement that “countries may still continue to apply withholding tax alongside the best practice.”[93]

Annex 3 Select bibliography

Addressing tax evasion and tax avoidance in developing countries, 2010, ITC, link.

African Tax Administration Forum, Business case, DFID, link.

Amendment to April 2013 MoU OECD-UK, 2013, OECD.

ATAF Regional Studies on Reform Priorities of African Tax Administrations, Africa-Wide Report, 2013, ATAF, link.

Automatic Exchange of Information: A Roadmap for Developing Country Participation, Final Report to the G20 Development Working Group, 2014, G20 Global Forum on Transparency and Exchange of Information, link.

Automatic Exchange of Information: An Opportunity for Developing Countries to Tackle Tax Evasion and
Corruption, 2014, Tax Justice Network, link.

Base Erosion, Profit Shifting and Developing Countries, IMF Working Paper 15/118, 2015, IMF, link.

Beyond BEPS: A Tax Policy Agenda for Developing Countries, 2015, ICTD.

Briefing on Base Erosion and Profit Shifting (BEPS) Implications for Developing Countries, 2014, Tax Justice Network, link.

Britain under pressure to end opposition to tax haven blacklist, 2016, The Guardian, link.

C20 Turkey Tax Justice Policy Paper, Summary of Policy Recommendations, 2015, Global Tax Justice, link.

Cayman’s Public registry of beneficial ownership: Smoke screen vs reality, 2015, Cayman Reporter, link.

Collect more spend Better – Achieving Development in an inclusive and Sustainable way, 2015, European
Commission, link.

EU releases world tax havens blacklist, 2015, EUbusiness, link.

Examples of successful DRM Reforms and the Role of International Co-operation, 2015, ITC, OECD, link.

Fifty Shades of Tax Dodging – The EU’s role in supporting an unjust global tax system, 2015, EURODAD, link.

Final JAHGA Paper, Annex III. Enabling Effective Exchange of Information: Availability and Reliability Standard, 2005, OECD, link.

Final List of Participants for the 6th Meeting of the Global Forum on Transparency and Exchange of Information for Tax Purposes, 21-22 November 2013, Jakarta, 2013, Global Forum on Transparency and Exchange of Information for Tax Purposes.

Financing For Development Conference, The Addis Tax Initiative Declaration, 2015, link.

First Monitoring Report of the Anti-Corruption Working Group to G20 Leaders on Individual and Collective Progress made by G20 Countries in the Implementation of the Seoul Action Plan, 2011, G20 Anti Corruption Working Group, link.

Frequently Asked Questions, 2013, Global Forum, link.

Fuest, Clemens, and Nadine Riedel, Tax Evasion, Tax Avoidance and Tax Expenditures in Developing Countries: A Review of the Literature, Oxford Centre for Business Taxation, 2009, link.

G20 Anti-Corruption Action Plan 2013 – 2014, 2013, G20, link.

G20 response to 2014 reports on base erosion and profit shifting and automatic exchange of tax information for developing economies. G20 Development Working Group, link.

Glossary of Tax Terms, OECD, link.

ICC urges consideration of broader trade implications of tax policies in response to BEPS recommendations, 2016, ICC, link.

International Centre for Tax and Development (ICTD), 2010, DFID, link.

International Manual, International Financial glossary, HMRC, 2016, link.

Issues in international taxation and the Role of the IMF, 2013, IMF, link.

Johannesen, N., Zucman, G., The End of Bank Secrecy? An Evaluation of the G20 Tax Haven Crackdown,
American Economic Journal: Economic Policy 2014, 6(1): 65-91, link.

Just for Show? – Reviewing G20 promises on Beneficial Ownership, 2015, Transparency International, link.

Mitigating Corruption Risk in the Extractives Sector: Examples from G20 members and other resource-rich economies, Background Paper, 2015, G20 Anti-Corruption Working Group.

Multi-year Action Plan on Development, Annex II, 2010, G20, link.

Options for Low Income Countries’ Effective and Efficient Use of Tax Incentives for Investment, 2015, IMF, OECD, UN, World Bank, link.

Outcomes ATAF Consultative Conference on New Rules of the Global Tax Age, 2014, ATAF, link.

Outcomes Document – ATAF Consultative Conference on New Rules of the Global Tax Agenda, 2014, ATAF.

Overall Evaluation of the G20/OECD Base Erosion and Profit Shifting (BEPS) Project, 2015, The BEPS Monitoring Group, link.

Part 1 of a Report to G20 Development Working Group on the Impact of BEPS in Low Income Countries, 2014, OECD, link.

Patching Up A Broken Tax System: Why BEPS is not the solution to poor countries’ tax problems, 2015, Action Aid, link.

Special Meeting of the OECD Task Force on Tax and Development on BEPS and Developing Countries and Summary of the BEPS consultations, Co-Chairs Statement, 2013, OECD Tax and Development Task Force, link.

Spillovers in International Corporate Taxation, IMF Policy Paper, 2014, IMF, link.

Submission of the United States of America to the United Nations Secretariat on Strengthening International Cooperation in Tax Matters, 2011, United States Mission to the UN, link.

Tacking tax evasion and avoidance, 2015, HMT, HMRC, link.

Tax Administration Diagnostic Assessment Tool, Business case, 2014, DFID, link.

Tax and Development, Principles to Enhance the Transparency and Governance of Tax Incentives for
Investment in Developing Countries, OECD, link.

Tax Annex to the St Petersburg G20 Leaders’ Declaration, 2013, G20, link.

Tax Competition in East Africa: A race to the bottom?, 2012, Tax Justice Network, Action Aid, link.

Tax havens told to drop opposition to UK call for central register, 2015, Houlder, V., Financial Times, link.

Tax in Developing Countries: Increasing Resources for Development. Fourth Report of Session 2012–13, 2012, IDC, link.

Tax Transparency Report 2015, 2015, OECD Global Forum on Transparency and Exchange of Information for Tax Purposes.

Tax Transparency, Business case, 2013, DFID, link.

Terms of Reference to Monitor and Review Progress Towards Transparency and Exchange of Information For Tax Purposes. Launch of a peer review process, 2010, Global Forum on Transparency and Exchange of Information for Tax Purposes, link.

The Future of UK Development Cooperation: Phase 2: Beyond Aid. Tenth Report of Session 2014–15, 2015, IDC, link.

Time to black-list the tax haven whitewash, 2011, Financial Times, link.

Transfer Pricing and the Investment Climate, 2015, World Bank.

Transfer Pricing Comparability Data and Developing Countries, 2014, OECD, link.

Transparency and Beneficial Ownership FATF Guidance, 2014, FATF/OECD, link.

UK aid: tackling global challenges in the national interest, 2015, HM Treasury, DFID, link.

UK wants direct access to Cayman company info, 2015, CNS Business, link.

Understanding the Tax Dispute: Heritage, Tullow and the Government of Uganda, 2011, ACODE Infosheet 16, link.

United Kingdom Position on “Strengthening Institutional Arrangements to promote International Cooperation in Tax Matters”, Including the Committee of experts on International Cooperation in Tax Matters, 2011, UK Mission to the UN, link.

Update on Voluntary Disclosure Programmes. A Pathway to tax compliance, 2015, OECD, link.

Annex 4 Methodology

The methodology for this review involved:

i. An assessment of DFID’s centrally managed programmes and non-spending activities. This included assessing DFID’s engagement with particular international processes and its work with other UK government departments. This assessment has involved a number of analytical approaches, including theory of change analysis, stakeholder analysis, process mapping and contribution analysis.

ii. A number of country case studies to examine DFID’s programmes and activities from the perspective of particular countries. These case studies involved an analysis of the country context and enabled us to incorporate developing country perspectives.

Because of the relatively narrow scope of this review, there was no need for a sampling strategy: all relevant aspects of DFID centrally managed aid programmes and non-spending activities were examined.

We have reviewed the full range of DFID and UK government activities, using two main data sources:

• Documentation: Documents included DFID commissioned reports, DFID programme documents and (especially for influencing activities) minutes from meetings, back to office reports, and email correspondence. Non-DFID sources included documents produced by other government departments, implementing partners, documents related to the selected geographical case studies, and other documents from academia, NGOs, media and others identified in the literature review.

• Key stakeholders: Key stakeholders included DFID, other UK government departments, DFID implementing partners (including the OECD), donors, international and southern NGOs, the private sector, academics and selected stakeholders from DFID partner countries (including government, tax revenue authorities and other donors). Different forms of consultations were used to collect views from stakeholders, including individual interviews, a roundtable with UKbased NGOs and the private sector, and a short survey aimed at DFID tax lead advisors.[94] The roundtable discussion with UK-based NGOs covered NGO participation in OECD-led processes and their advocacy on international tax. The roundtable discussion with the private sector covered the response by multinationals to international tax standards and regulations they need to comply with.

The following analytical approaches were used to inform our findings:

1. Theory of Change analysis: The theories of change guiding DFID’s support to International Tax were extracted from DFID documentation. Complementary aspects of DFID’s work, implicitly or explicitly captured in those programmes, such as policy dialogue and cross-government working, were also analysed. Based on detailed discussions with DFID and the findings of the literature review, we have analysed these theories of change to determine the extent to which the key assumptions were made explicit, highlighting any gaps or inconsistencies (see Figure 5).

2. Stakeholder analysis: Analysis was carried out to characterise DFID’s partnership with other UK government departments and inform the review findings on cross-government working. This included a mapping of UK government positions on international processes related to cross-border tax avoidance and tax evasion, their coherence with DFID’s policy objectives, and, an in-depth review of working relations between DFID, Treasury and HMRC.

3. Process mapping: International processes were mapped out and analysed using a series of timelines, to identify sequencing, identify DFID’s entry points and analyse the main outputs. A framework was developed to collect data relating to DFID’s approach to influencing, covering its choice of issues (what), its activities and inputs (how), the timeline of events (when), the choice of partners (who) and the linkages with developing countries (where).

4. Contribution analysis: To assess DFID’s contribution to international processes and cross-government working, this review has used a range of tools, including: sequencing (eg time correlation between DFID activities and decision-making), strategic consistency (eg clarity and consistency of policy engagement), quality of dialogue (eg frequency, inclusiveness), interactions (eg convergence/divergence of views) and counterfactual considerations.[95]