The most lively debate in the first two days of the UN tax committee meeting ended with the decision to start work on a new article for the UN’s model tax treaty that would allow developing countries to levy a tax on payments made to overseas providers of ‘technical services’. The advocates of this position support the view set out in the input paper [pdf] produced by Canadian tax professor Brian Arnold, which explains the problem as follows:
The payments for technical services erode the source country’s tax base, but such payments are often not taxable by the source country under the provisions of the United Nations Model Convention treaty. As a result, multinational enterprises sometimes use fees for technical, management and consulting services to strip the profits of their subsidiaries.
In the case of SABMiller, the brewer accused of dodging taxes [pdf], a Ghanaian company made significant payments to its Swiss sister for such technical services. Ghana, like some other developing countries, had been clever enough to ensure that its tax treaty with Switzerland still allowed it to tax these payments, but to do so it had been forced to ask for the inclusion of a provision within this treaty that was not based on anything in the UN or OECD model conventions that underpin treaty negotiations.
At first sight, then, there’s a simple institutional reason for adding this provision to the UN model: over 100 treaties now include something to this effect, but there is no standard way of doing it. And if there’s one thing international tax practitioners say they dislike, its this kind of anarchy, which they argue makes life harder for the treaty negotiators, means the provisions adopted tend to be less well thought-out than those arising from debate and negotiation inside the UN committee (or its OECD equivalent), and makes life more complicated for multinational taxpayers, who have to check the provisions of each treaty to be sure they’ve got it right.
Yet when it came to a vote, a number of countries and committee members, predominantly those from the OECD, opposed the move. To understand their arguments, we have to look a bit more at the reasons that others support the change.
As noted above, one reason is to provide a tool to reduce the impact of international tax dodging techniques. Enforcing the complex ‘transfer pricing’ rules needed to verify whether service fee payments are appropriate is hard, too hard for many developing countries at the moment. Being allowed to levy a ‘withholding’ tax on the payment itself – which will usually be offset by the home state of the company receiving the payment through a tax credit – is a first line of defence that reduces their reliance on transfer pricing enforcement for protection. This was the line of reasoning set out by many developing country delegates yesterday.
In contrast, the US and UK official observers used phrases such as “such profit shifting, if it exists…” to cast doubt on this justification. According to these countries, technical service payments aren’t anything to do with tax dodging. Instead, they allow the international tax system to work as it should do, giving countries that specialise in high value services, which contribute a lot to a multinational’s profits, a decent share of the right to tax those profits.
This brings us to the second reason for the proposed change, which is about a much more fundamental change in the global economy: the growing irrelevance of physical presence in understanding the distribution of economic activity.
Under the model treaties, a developing country may only tax a foreign service provider (such as construction companies or management consultants) if it has a ‘permanent establishment’ in the country. Generally this means they must have an office or other physical presence, or some employees, in the country for more than half the days in a one year period, or else a “fixed base” that they use regularly. There are always exceptions, of course, and here’s an interesting one: there are specific exemptions for highly-paid sports players and performers, which a country can tax even if they stay just a few hours or days for a match or concert.
The UN committee’s proposal, as characterised by the UK’s official observer, crosses a significant threshold. It allows a country to tax the income of a service provider even if it has no physical presence in their country. One example given was that if a CCTV camera installed in the UK is monitored by employees of a company in India, under the new proposal the UK might have the right to levy a tax on the payment from the British client to that Indian service provider.
Another way of understanding the inconsistency is to compare the situation with royalty fees. If the CCTV system runs on technology developed by the Indian company, and the British company pays a royalty fee to use it, the UK would be already allowed to tax that payment. But if someone comes over from India to help install the system, at present the UK can’t tax the fee paid for this, so long as the visit is only temporary. Under the UN’s proposal, both payments could be taxed by the UK.
I’ve used the UK as an example, but the UK exports lots of the kinds of services that will be affected, so it won’t gain from the UN proposal, and would lose a bit because it gives British businesses a tax credit where they have had to pay these withholding taxes overseas. Developing countries, which import such services, would be the winners, which is why it’s more than just a technical fix – it’s a fundamental shift in the distribution of taxing rights from rich to poor countries.
And yet the answer to the question in the title of this post is probably no. Developing countries have already decided what’s in their interests, and in a piecemeal way they’ve started to make it a reality. Provisions on technical services don’t just appear in treaties between relatively small countries such as Ghana and Switzerland – there is also one in, for example, the US-India tax treaty. One might imagine that the decision to include it depends nonetheless on the power dynamics of particular negotiations, and in that context an updated UN model would strengthen developing countries’ hands. But it is hard to see how it would be crossing a threshold that has already been crossed by many UN member countries, including some of the most powerful.