Again Agency PEs (but this time Secret Agents)

I am back from vacation (more detail in some days). Meanwhile, in the eve of the 2006 IFA Congress in Amsterdam, where the topic of the Attribution of Profits to a Permanent Establishment will be thoroughly discussed, the highly recognized Richard J. Vann made available a paper on the (secret) topic of agency permanent establishment (PE) under tax treaties. As you may have noticed (by the old enteries) this topic is dear to me and it is with pleasure that I post the link to the paper at the same time I will print it for the first time and read it! Probably the readers of Talk Tax will expect a comment on the paper in the coming week and I will do the utmost to fulfil such expectation.

In the meantime, here is the abstract to the paper Tax Treaties: The Secret Agent's Secrets also published in the last number of the British Tax Review (No. 3, p. 345, 2006)

Abstract:
Recently two apparent paradoxes have been revealed about an agency permanent establishment (PE) under tax treaties, first that it is possible to avoid an agency PE by exploiting a difference between civil law and common law on agency (often referred to as commissionnaire structures) and secondly that if the agent is rewarded with a market-value fee there will be no profits to attribute to an agency PE. This article demonstrates that these problems have been present since the origin of tax treaties, and that they stem from a form-over-substance approach to PE tests which relegates the independence test in defining PEs to a secondary role, from the use of a different indeterminate independence test in transfer-pricing rules and from the definition of the firm in terms of common ownership. Underlying the problems have been inconsistent views on how value is generated within a firm. The solution is to settle on a workable theory of value, to apply substance over form, and to realise the significance of independence in defining the boundary of the firm. Under current treaty law, the second asserted paradox is not correct if treaties are interpreted in a sensible manner.