Speech

Joint Conference by the IMF's Fiscal Affairs Department and European Commission's Directorate General for Taxation and Customs Union
Brussels, February 23, 2015

As prepared for delivery

It is a great pleasure for us in the Fiscal Affairs Department (FAD) to team up with European Commission's Directorate General for Taxation and Customs Union (TAXUD) in organizing this conference. It is particularly gratifying for me since, just before joining the IMF, I worked very intensely with TAXUD on Taxation of the Digital Economy.¹

We have worked closely with TAXUD on a range of topics over the years. Today marks another step in our close cooperation, which I am sure will continue and intensify in the coming years.

Certainly, today’s topic is one on which FAD and TAXUD have developed views in close cooperation. We both see “debt bias” as a central tax policy issue in many countries. It is an issue in need of serious analysis and action. And we both believe that bringing together key experts in the field to exchange views and experiences can make a real contribution to much-needed progress in the area.

Let me give a brief overview of how thinking and practice have evolved in the area of “debt bias” as context for today’s discussions.

The IMF and the public finance community more generally have long recognized that tax distortions encourage businesses and households to finance their activities by debt rather than equity. This clearly creates inefficiencies and deadweight loss. But for many years this attracted little attention. It was not even clear how large the deadweight costs were. But this has changed in the last few years. Why?

This strong interest is first and foremost a consequence of the crisis. A crisis that started when I was working, here in Brussels, as head of the Bureau of European Advisers, between 2007 and 2010. I remember that, in 2007, and most of 2008 many held the view that Europe would be sheltered from developments elsewhere. “Decoupling” was a frequently used word. Facts have not been kind to this view.

In the field of today’s conference, the Great Global Crisis justified an increased concern that tax incentives to excess leverage that can intensify crises and perhaps even make them more likely.

These issues were stressed in the IMF board paper on Taxation and the Crisis in 2009.² The paper concluded that debt bias at both corporate and personal levels was not a proximate cause of the crisis, but it certainly did not help. Analysis in TAXUD reached the same conclusion.

The second reason for increased attention on the debt bias issue has been the focus on interest deductibility as a major tool in multinationals’ armory of tax avoidance tools: With increased globalization, multinationals are increasingly able to rearrange their financial affairs so as to minimize their tax bills. This is particularly important for public revenue in countries with high corporate income tax rates—not least developing countries.

The IMF Tax Spillover Paper in 2014³ emphasized debt shifting as one of the important channels of tax planning. Interest allocation rules have also been one of the OECD’s Base Erosion and Profit Shifting (BEPS) action areas, and we look forward to hearing more on that later.

These two issues of debt shifting and interest allocation rules are inter-related but are not the same. When one entity within a corporate group lends to another, for instance, it may serve to reduce the groups’ total tax payments. But such internal debt does not in itself affect the overall riskiness of the group’s financial position—and so does not pose the same risk to financial stability as money borrowed from outside the group.

Let me focus on the financial stability aspect of debt bias. One lesson the crisis taught us is that the social cost of excessive leverage may be especially large in the financial sector. We saw that the failure of systemically important institutions led to two outcomes: either the unmitigated collapse of the institution resulting in high external costs to the rest of the economy or large bail outs that placed significant fiscal strain on the public sector.

The implications of this were pursued in the Fund’s G20 report on Financial Sector Taxation in 2010. ⁴ That was largely motivated by the question of whether special bank taxes were appropriate in the aftermath of the crisis. And one of the key points made in the report was that before thinking about new taxes it would be good to fix problems in the existing system—among which debt bias was prominent.

This report was for us the start of a continuing program of work. Let me mention two.

• The first is analytical work by staff in the Fiscal Affairs Department on theimpact of corporate tax bias on bank leverage in a series of empirical studies using micro data for banks.⁵ The authors come to the important conclusion that an average bank is as responsive as any other firm—so tax effects on leverage are not overwhelmed by regulatory constraints. They also found that large banks are much less responsive to taxation. But not too much comfort can be taken from this conclusion. Small changes in the leverage of large banks can lead to notable changes in the probability of their failing which can unleash huge social costs.
  
  
• The second is the recent book Taxation and Regulation of the Financial Sector (December 2014).⁶ This book was edited jointly by staff from FAD and TAXUD. It covers a wide range of issues at the interface between finance and public finance. I believe that this is a massively under-researched area. For instance, why do countries combine capital requirements for banks that encourage equity finance while providing tax incentives to do the exact opposite? Are bank levies better than direct regulation as a mean of addressing externalities associated with bank failure?

Addressing debt bias is now firmly on the policy agenda in many countries. It perhaps ought to be on the agenda in many more. Belgium and Italy have gone furthest, and I look forward to learning more about their experiences today. Others are also considering steps to address debt bias by reforming their tax system—Norway and Sweden are recent examples.

Many European countries have over the years imposed restrictions on interest deductibility. But these were mainly intended as anti-avoidance rules against international debt shifting. Indeed, they are not comprehensive and bring along a lot of complexity.
And many countries have introduced bank levies as part of their response to the crisis. This can be seen as partial attempts to address debt bias among banks. These do indeed seem to have had some effect.

And, as we just heard, the European Commission pays ample attention to the issue in its recommendations to Member States.

Policymakers are indeed showing increased interest in comprehensive approaches to the problem, which had seemed before to excite only academics. Prominent among these are:

• One: the Comprehensive Business Income Tax which eliminates the deduction for interest.
  
  
• Two: providing an Allowance for Corporate Equity (or variants thereof).⁷ Such an allowance was introduced in Belgium and Italy, and it was recommended for the UK by the Mirrlees Review Committee.

The policy agenda is rich and full, with many questions for us to discuss during our deliberations today and tomorrow. Is Allowance for Corporate Equity, for instance, the best way to tackle debt bias, as we in FAD have tended to believe? Can it be designed to minimize any adverse revenue impact? And much remains to be learnt about the impact of debt bais in the financial sector. How, for instance, do taxes and regulatory provisions shape the structure of financial institutions, for example in the organization of branches versus subsidiaries and where those are located?

The topic of this conference is extremely timely and relevant. Timely because of the increased recognition that something needs to be done to reduce possible distortions. And relevant because the economic implications at stake are large, potentially very large.

The issue of debt bias is high on the agenda of policymakers and international organizations alike. For instance, the G20 Finance Ministers earlier this month asked the International Organizations to draft a report on the issue for discussion during their September meeting. The IMF Board of Directors has also asked FAD to extend its analysis on the relationship between tax policy and financial stability.

This conference offers a unique opportunity to bring together policy makers and academia to take stock of what we know and discuss recent country experiences with actual policy reforms. For instance, I very much look forward to hear more about the experiences with Allowance for Corporate Equity regimes in Italy and Belgium; and also to learn more about the experiences with restrictions to interest deductibility in Germany and the US and discussions in the G20-BEPS process.

Understanding how taxation and regulation influence financial stability remains an important intellectual challenge. This has significant implications for policy design. This is an area in economics that is still underdeveloped. Many questions remain unanswered. I eagerly look forward to the conference’s contributions and discussions on the financial sector. I am sure they will inspire and help us in our future work.

Thank you.

---

¹ TAXUD on Taxation of the Digital Economy

² http://www.imf.org/external/pp/longres.aspx?id=4339

³ http://www.imf.org/external/pp/longres.aspx?id=4873

⁴ http://www.imf.org/external/np/g20/pdf/062710b.pdf

⁵ http://www.imf.org/external/pubs/cat/longres.aspx?sk=25727.0 (forthcoming in JMCB) and http://www.imf.org/external/pubs/cat/longres.aspx?sk=40341.0

⁶ http://mitpress.mit.edu/books/taxation-and-regulation-financial-sector

⁷ See for instance de Mooij (2011); http://www.imf.org/external/pubs/cat/longres.aspx?sk=24810.0