While Lee Sheppard provides a typically thoughtful and provocative analysis of transfer pricing, debating transfer pricing versus formulary apportionment feels much like rearranging deck chairs on the Titanic. The real problem with international tax lies not in which technical rules should apply, but with a more fundamental question of how to divide the international tax base.
The original idea behind transfer pricing was to mitigate multiple countries taxing the same income - the so-called "double tax" problem. What is often forgotten is that transfer pricing has done a pretty good job at mitigating double tax. Unfortunately, it did so at the cost of creating double non-taxation, or income falling through the cracks. Absent international agreement, there is no way to reduce one without increasing the other. In other words, fixing double non-tax could just bring back double tax. That may well be better than what we have, but it is definitely not free.
I agree with Sheppard that there are numerous problems with transfer pricing. I also agree that a shift to formulary apportionment would recapture much of the tax base lost to transfer pricing. This has to be true. For example, if tax base is allocated to the country of consumption, high consumption countries will benefit. But there is little reason to believe low consumption countries would go along.
So how to get countries with disparate interests to agree on anything when it comes to division of tax base, especially when some countries can't even agree on signing a tax treaty with each other? Thinking outside the tax treaty could well provide some answers. This is not necessarily mutually exclusive with a move to formulary apportionment, but it may be necessary to fulfill its promise.