The IMF seeks an alternative to the arm’s-length principle
Ruud de Mooij, who oversees tax policy at the IMF, talks to TP Week about the future of global tax and why the arm’s-length principle may no longer be fit for purpose.
Despite its fiscally conservative reputation, the IMF appears to have taken a radical turn on the arm’s-length principle (ALP) in its report on corporate tax. The rise of new technologies and the online economy might make traditional standards null in
IMF Managing Director Christine Lagarde framed the possible solutions in terms of a choice: Either create a minimum tax rate within the existing system or redesign the entire architecture of international tax. This is where the debate gets controversial.
The IMF is weighing up the advantages of global unitary taxation against a destination-based cash flow model. The paper shies away from endorsing any one policy, but it’s clear that this a game-changer.
Josh White: How did the IMF arrive at the view that the international corporate tax system needs to be overhauled?
Ruud de Mooij: From the analytical work, we have learned how big the spillovers of domestic tax policies are on other countries, through BEPS
JW: One of the most important points in the report is that the arm’s-length principle might not be fit for purpose. Why do you think some people have been surprised that the IMF would take a critical stance on the ALP?
RM: The ALP seems to work fine for some transactions, but certainly not all. The way in which arm’s-length prices are set for, for example, intangible assets or risk is incomprehensible for an economist like me, even conceptually. And many transfer pricing experts I speak to acknowledge that the fiction of the ALP is to a very large degree arbitrary and easy to manipulate. These issues have only grown in importance recently.
Our analytical work shows, for instance, how large the implications for government revenues in developing countries are; and our staff working in the countries experience this every day in their work. My biggest concern is that the system is so problematic for the developing countries. Of course, I understand that many people make a living out of the ALP and that change isn’t easy; but views on this seem to be rapidly changing.
JW: Although the report does not endorse any one proposal, it does explore different options and one of the most interesting is the destination-based cash flow tax (DBCFT). What are the strengths of the DBCFT in overcoming the limits of the international tax system?
RM: The IMF paper does not endorse any one proposal as each has its distinct merits and drawbacks, for example, regarding how they mitigate tax spillovers, their legal and administrative feasibility and the impact on developing countries. For the DBCFT, both components, destination-based (DB), and cash flow tax (CFT) have some key economic attractions. Compared to the current corporate income tax, for instance, a CFT eliminates distortions on investment and removes the bias for companies to finance their investment excessively by debt. This has important economic benefits.
The DB, moreover, provides a complete solution to profit shifting – which is no longer possible – and tax competition as low tax rates no longer attract tax base. The spillovers that undermine the international tax framework are thus drastically reduced, as the company’s consumers are less mobile than its location of source or residence. This is not to say that there will be no challenges with the DBCFT.
For instance, its benefits will only be fully
JW: There has been some intense competition between the OECD and the EU on finding a solution to digital tax. To what extent is the DBCFT a possible solution to the problem of digital taxation?
RM: If you are referring to the problem of perceived under-taxation of highly digitalised businesses due to profit shifting through mispricing of intangible assets, then yes the DBCFT entirely eliminates that problem due to the border adjustment of taxes. But, DBCFT also addresses another issue with digitalisation, namely that people find it unfair that highly digitalised businesses currently pay no tax in the country where the users or consumers of digital services reside. In that sense, the DBCFT is probably the most comprehensive solution to the various challenges associated with digitalisation.
JW: Critics of the DBCFT claim that it is just another form of VAT and it will fall hard on the heads of the poor. Do you think this is a fair criticism? If so, how might the proposal be made less regressive?
RM: There are many misconceptions about the DBCFT and it’s important for a good debate that these are straightened out – whether you like the idea of a DBCFT or not. Clearly, the DBCFT is not a VAT because labour costs are deductible from the base of the DBCFT. This is a big difference. It also means that its incidence is different from that of a VAT. On this, there is still some debate ongoing between analysts, but it seems most plausible that the DBCFT will be borne by the ultimate shareholders of the firms and thus be more likely progressive than regressive.
Moreover, empirical analysis by some of my IMF colleagues finds that low-income countries will likely gain from a DBCFT compared to the current system. This is for two reasons: first, they no longer lose revenue from profit shifting; and second, they consume out of development aid so that the destination base is broader than the
JW: Many tax policymakers and reformist NGOs think some kind of formulary apportionment would be a far better solution, but they don’t agree on the fine details. Does the IMF have a view on how formulary apportionment would work best in practice?
RM: Formulary apportionment has several merits, but also raises several new challenges, as discussed in the IMF paper. One is the large redistribution that can occur. To mitigate this, the IMF report discusses another idea of ‘residual profit split’. Under such a system, the income of a multinational company would be split into a ‘routine’ return on investment, and a ‘residual’ return that is the excess over normal returns.
The ‘normal’ return would be allocated to source countries, potentially by pricing ‘routine’ activities based on the ALP. The residual profit would be shared on a formulaic basis, which avoids problems with ALP where they are often most severe.
The weights in the formula could be based on the destination of sales, which would have the merit of limiting tax competition and be closer to income attribution to countries where the users and consumers of digital services reside.
JW: What is the next on the agenda for the IMF following the publication of this report? Will the
RM: The IMF is not a standard-setting body, like the OECD’s Inclusive Framework or the UN Committee of Experts. Our role has always been and continues to be twofold.
First, we continue to support our member countries with analysis and technical advice on tax issues. The extensive experience from working in developing countries, moreover, provides a perspective that we can bring to the attention of standard setters, for instance through our role as observers and our collaboration with the OECD, the World Bank and the UN in the Platform for Collaboration on Tax. We think that it is indeed essential that the interests of developing countries are heard and properly reflected in future reforms of the international tax framework.
Second, we continue to contribute to the global debate by providing independent economic analysis and assessments. Heated technical discussions on tax matters can often easily overlook what the reforms ultimately aim to achieve or what its ultimate economic implications are. The IMF has the responsibility to inform the public debate on such impacts.