Although their application is wider, the single example is basically as follows:
Example. USP owns US1, which owns FT with no gain or loss in the FT stock. USP also owns FS, which has lots of earnings and profits. FS buys USP stock in the market for $100 and swaps it with US1 for the stock of FT. Section 367(a) applies to make this a recognition exchange as to US1, rather than a nonrecognition B reorganization. However, the new regulation applies to instead treat FS as distributing $100 to USP, which USP then contributes to FS. US1 enjoys nonrecognition treatment (for its zero gain) under section 354 because its stock exchange is treated as a B reorganization.
What Is Going On?
The saga of these Killer B regulations is a long one, and the tale has been told many times. It is highly unlikely that taxpayers will stumble into this regulation unawares, therefore the principal message is NOT TO DO A KILLER B unless you have very special facts (not only no gain in the FT stock, but no E&P in FS, in which case you don't need to use a Killer B to repatriate the $100 cash).
What is more interesting is to (a) evaluate how Treasury came to this particular exercise of its regulatory powers and (b) see if the regulation could accidently impact less obviously "killer" transactions.
If you didn't know anything about the regulation and its history and were told about the example above, you would think it runs afoul of Section 956. Section 951(a)(1)(B) requires a U.S. shareholder (which would be USP) to include the Section 956 amount, which is a controlled foreign corporation's (that would be FS) investment in U.S. property, which includes stock in a corporation such as USP. The example above shows that Treasury is most worried about the use of the earnings and profits of FS for the purposes of USP, such as redeeming its shareholders' stock. That is also the worry of Section 956.
But there is a problem with applying Section 956: it only operates on the U.S. property owned by the controlled foreign corporation at the end of a quarter, which on the facts of the example can be zero, because FS does not receive and hold the stock of USP; instead, it swaps it to US1 for the stock of a foreign corporation. Treasury might have the regulatory authority to avoid this limitation under Section 956(e): "The Secretary shall prescribe such regulations as may be necessary to carry out the purposes of this section, including regulations to prevent the avoidance of the provisions of this section through reorganizations or otherwise." It seems that FS has avoided the provisions of Section 956 through a reorganization: it has reorganized away the acquisition currency, the stock of USP, which otherwise would have been U.S. property held at the end of a quarter.
However, Treasury chose to deal with the situation under Section 367. That required two regulations. First, Section 367(a) normally would apply to cause recognition to US1 because it is a U.S. person transferring property to a foreign corporation in a transaction to which Section 354 otherwise would apply. The revised Reg. § 1.367(a)-3 allows Section 367(a) to turn off nonrecognition treatment in these cases unless US1's gain is less than USP's dividend and gain on a deemed distribution of the $100 from FS. Second, to deal with the excepted cases, Treasury created Reg. § 1.367(b)-10. It contains the example summarized above, taxing USP on a deemed distribution of $100.
Section 367(a) is basically about the prevention of the movement of gain offshore without recognition by a U.S. taxpayer. Section 367(b) is basically about preserving the ability to tax U.S. taxpayers on the earnings of foreign corporations when repatriated. The Killer B regulations apply both: if US1's gain is greater than the deemed dividend, then Section 367(a) applies; if the deemed dividend would be larger, then a repatriation is created and taxed. Nevertheless, the example most closely resembles an investment of the earnings and profits of a controlled foreign corporation in U.S. property. It is not clear that adding to the already over-convoluted Section 367 regulations was the most appropriate way to deal with this problem. If Treasury did not feel it had the regulatory authority under Section 956(e), then one wonders whether it has the regulatory authority under the less directly implicated sections of the code.
No one would walk into the repatriation scheme in the example above without being advertent to the Killer B regulation. However, taxpayers need to know that the regulation also can apply if the parent is foreign and the acquiring subsidiary is domestic, and any outbound dividend to the parent would be taxable to some extent after application of treaty exemptions.
Example. FP owns US1, which owns FT with no gain or loss in the FT stock. FP also owns US2, which has lots of earnings and profits. US2 buys FP stock in the market for $100 and swaps it with US1 for the stock of FT. Section 367(a) does not apply. However, the new regulation applies to instead treat US2 as distributing $100 to FP, which FP then contributes to US2. US1 enjoys nonrecognition treatment (for its zero gain) under Section 354 and the stock exchange is treated as a B reorganization. But withholding tax can apply to the deemed dividend.
The outbound example above explains why Treasury could not depend wholly on Section 956 to solve the Killer B problem, because it does not involve an investment by a CFC in U.S. property. However, the outbound example also illustrates the real unifying theme of Treasury's concern: disguised distribution by a subsidiary to a parent by purchasing parent's stock either from the parent or from its shareholders. But if Treasury had addressed the problem as one of disguised distribution in a limited set of cases of purchasing parent stock, it would have had to deal with the lack of guidance on that subject in purely domestic cases. Therefore, it chose, as usual, to create special rules for foreign cases that perhaps are not all that unique.