In Come 2012 Taxes: GE’s tax shelter isn’t really, says appeals court

January 25, 2012

Tax law

(Editor’s note: With January comes the start of tax season.  Throughout the month, we’ll be looking at current and possible future developments in tax law.)

Click here for the first post on U.S. v. Home Concrete and Supply.

Click here for the second post on FACTA’s new reporting requirements.

Click here for the third post on the IRS’s bait-and-switch.

Yesterday, the U.S. Court of Appeals for the Second Circuit ruled that a tax shelter used by General Electric’s subsidiary “TIFD III-E Inc” wasn’t really a very good one (i.e. it didn’t actually shelter income from tax liability).

I’m not going to pretend that I actually found the opinion’s discussion on tax law interesting, but I did learn a little something about tax shelters, which, previously, I had known very little details about.

This particular tax shelter, called “Castle Harbour LLC,” (the name doesn’t seem to make an attempt to hide its purpose as a shelter), is a partnership between TIFD III-E and two Dutch Banks.

The two banks, ING Bank and Rabo Merchant Bank, are, being foreign entities, not subject to U.S. tax liability.

GE avoided tax liability by having the partnership’s income transferred from TIFD III–E to the banks.

In practice, $310 million of taxable income was shifted to the banks from 1993 to 1998, which saved GE $62 million in taxes (there were actually several different mechanisms employed to achieve this effect, but you wouldn’t still be reading this if I had actually detailed them all).

Since partnerships are distinct legal entities, and because of how partnerships operate, the transferring of TIFD III–E’s income to its bank partners was like someone’s right hand giving money to his left hand.

In this case, the right hand was responsible for paying U.S. income taxes, while the left hand was not.

As mentioned above, the appeals court didn’t agree that this should have worked.

The reasoning for this is complex and rooted in specific interpretations of tax code provisions, and so, unsurprisingly, it can get very tedious.

Really, though, it can be summed up fairly simply (kind of like the rest of this case has been so far, albeit somewhat crudely).

The appeals court navigated through the “maze of contractual provisions in the partnership agreement” to essentially determine whether or not the Dutch banks were, in fact, partners.

The conclusion?

They were lenders more than partners.

To be a bona fide partner in a partnership, you must share in the partnership’s profits and losses based on your ownership share (i.e. if the partnership profits $100, and you have a 25% share, you profit $25).

Despite the inclusion of language and provisions in the partnership agreement to appear otherwise, the court found that the Dutch banks were guaranteed an annual return of about 9% on their $117.5 million.

That makes them investors, not partners, so the hand-passing analogy mentioned earlier doesn’t apply.

Granted, since I don’t really know a lot about tax shelters, I don’t know if this particular method has been employed widely, so I don’t know how wide of an impact this ruling will have.

However, this ruling does seem to be consistent with a trend uncovered by the other tax law posts this month: the federal government is getting serious about tax enforcement.

Assuming that this decision wasn’t influenced as part of a concerted effort across all three branches of government toward that end, it nonetheless comes at a time when both the IRS and Congress are looking to increase revenue through stricter enforcement.