Why arm’s length pricing does not work

Posted on

If those commenting on this blog were to be believed (and they should not be) the OECD's preferred method of allocating profits between states allocates 100% of group profits to jurisdictions and no more or less, and does so in an environment of mutual cooperation where outcomes are assured and fair tax is paid.

That's not true. As Patrick Love has written on the OECD's own blog:

There is no simple method for calculating a transfer price, so the final value is the result of a negotiation between the company and the tax authority. Ideally, this would be based on equal access to information, a shared objective and a “zero sum game” where an exemption in one jurisdiction is offset by tax in another. It’s not. International business consultancies have more people working on transfer pricing than any national tax authority. Prem Sikka of Essex Business School, co-author of a paper on The Dark Side of Transfer Pricing, claims that “Ernst & Young alone employs over 900 professionals to sell transfer pricing schemes. The US tax authorities employ about 500 full-time inspectors to pursue transfer pricing issues and Kenya can only afford between three and five tax investigators for the whole country.

So let's be clear about why transfer pricing is not the neat zero-sum game its apologists would like to suggest it is.

First, there is no guarantee a company will send matching information to each tax authority.

Second, there is no reason why the tax authorities need agree.

Third, the tax authorities may not know they disagree and there is no obligation on a taxpayer to tell them if that outcome is in their favour, as it can be.

Fourth, there are insufficient resources, as noted above giving asymmetric power to companies.

Fifth, there are insufficient double tax agreements. Developing countries have few and there are almost none with tax havens. by definition then there is no reciprocity in much of the system.

Sixth, as I have noted, all transfer pricing adjustments are to a wholly artificial standard - intended to tax the profit that would arise if companies were not related, but as a matter of fact they are. This results in under-taxation because companies are only related and under common control because more money is made as a result - but that excess is not taxed under OECD rules - meaning some profit always falls out of tax under existing TP rules.

I could go on - but you get my point. Current OECD transfer pricing rules 1) invariably under tax and 2) do not result in reciprocal tax adjustments. Those claiming otherwise are a) not aware of the facts b) misrepresenting the truth c) deliberately misinforming for the promotion of self interest. I am not sure what else I could reasonably conclude from any such claim. (a) is of course quite common. I am afraid I hear (b) and (c) too often.