I do a lot of research on tax treaties, tax treaty information exchange, and especially competent authority negotiations under the mutual agreement procedure (MAP)--i.e., the technical means by which countries allocate tax revenues amongst themselves. Don't stop reading, this is an interesting story, I promise.
One of the things I study about what the competent authorities are doing under MAP is just how secretive the whole structure is. It seems like it took a Tax Analysts FOIA battle to get the IRS to start releasing even the most basic of data on its competent authority program, and then it's only done so in aggregate statistical form. If you want to know why I think its important that we know what it is the competent authorities are doing and why the IRS ought to be telling us at least a little bit about this program, you can read my forthcoming article called How Nations Share, here is a draft.
But this is not about those ideas. Instead, this is about an annoying occurrence that has now come up two years in a row with respect to IRS information dissemination about the competent authority/MAP Program. This is what happened:
I was searching around online looking for anything to do with competent authority cases, and came across this little newsletter by Ernst & Young, dated January 21, 2010. Interesting! It tells me that the IRS just released the "2009 Competent Authority Statistics Report on 15 January 2010," and then it describes the report, complete with a number of charts, graphs, etc. This is a fascinating secondary source but of course within two seconds of realizing what it was about, I wanted to see the primary source. I check the E&Y document but there is, annoyingly, no helpful ink, not even a spelled out URL, nothing but the name of the report and its purported release date. So I google, and dig around, and google some more, and try everything I can come up with to find this elusive 2009 report. And where is it?
No where. No where at all.
Yet E&Y must have it, so how can this be? I asked my local rock star law librarian for help and she made inquiries to the IRS. Finally, at least a month later, my librarian comes back to me with an email direct from the IRS, with an attachment--a Word document showing those statistics E&Y used, accompanied by the statement "I believe you requested this information. Let me know if it is not what you wanted." it is! Thank you, IRS! But...
OK, that's odd, I thought. Why would the IRS go to all the trouble of putting together these statistics, and not release them to the public, like in a press release, such as on the IRS website, for example, at the IRS newsroom? Further, if this document isn't released to the public, then how come E&Y has it, and has it in time to produce a nice glossy pamphlet (ok, it's an online pdf but my screen makes it look glossy) on the topic within six days of its purported "release"?
I dismissed my thoughts about inner circles, agency capture, conspiracies, and the tax elite, in order to get on with the business of analyzing the data.
Now I am starting to think that the "release" IS the story.
Because here, a year later, is again a very nice E&Y document, dated January 19, 2012, that looks very much like last year's. E&Y may be going a bit slower--this time the report has been issed almost month after the purported "release" of the latest competent authority statistics, on December 16, 2011.
And again the actual report is nowhere to be found.
Why is this? I've again asked my librarian to investigate, but what on earth is going on here. What possible purpose could be served by "releasing" data but only actually furnishing it to E&Y? Actually, it seems KPMG might have got their hands on it this year, as well, but didn't bother with it too much.
Any thoughts on the elusive competent authority report? And if you have a copy of the report, can you send it to me?
And while we're on the subject, just exactly how long has the IRS been compiling and not-releasing competent authority statistics?
Addendum: Since at least 1994, by releasing only by "press drops," and it is maddening to say the least to get any kind of complete dataset. More to come on this subject.