ArticlesVol. 8, No. 3

A Framework for Testing Regulatory Authority

Citation:8 Colum. J. Tax L. 326


In FDA v. Brown & Williamson Tobacco Corp.,[1] the Supreme Court considered whether Congress had granted the FDA authority to regulate tobacco and tobacco products.  The Court held, in a 5-4 decision, that Congress had not granted such authority.  Writing in dissent, Justice Breyer primarily relied on the text of the Food, Drug, and Cosmetic Act (FDCA).  The FDCA granted authority to the FDA to regulate “drugs” and “devices.”[2]  Nicotine is a drug; a cigarette is a device to deliver nicotine.  Justice Breyer wrote: “In short, I believe that the most important indicia of statutory meaning—language and purpose—along with the FDCA’s legislative history . . . are sufficient to establish that the FDA has authority to regulate tobacco.” [3]

In an opinion written by Justice O’Connor and joined by, among others, Justices Scalia and Thomas, the majority looked beyond the statutory language “in isolation” and held that, after examining the FDCA as a whole, the FDA’s own statements on the limits of its jurisdiction, and 35 years of tobacco-specific legislation (as well as legislative history and bills proposed but not adopted), “it is clear that Congress intended to exclude tobacco products from the FDA’s jurisdiction.”[4]  In short, notwithstanding that nicotine is a drug and a cigarette is a drug delivery device (at least in the normal sense of those words), that’s not what Congress meant.

When the Treasury Department and the IRS published proposed and temporary regulations under Sections 385,[5] 956, 7701(l) and 7874, many tax practitioners had a reaction similar to the Court’s reaction in Brown & Williamson, particularly to the proposed Section 385 regulations (the “Proposed 385 Regulations”), and particularly to Proposed Treasury Regulation Section 1.385-3 (the “per se rule”): whatever the statute and the legislative history of Section 385 can be read to say, that’s not what Congress meant.[6]

And yet, the initial reaction was that the Proposed 385 Regulations were clearly within the scope of authority granted to the IRS under Section 385.[7]  “The explicit language of Section 385 gives the Treasury secretary direct and powerful regulatory authority to reclassify debt as equity and thereby transform a deductible interest payment into a nondeductible dividend.”[8]

That may be the case, but to assess whether the statute authorizes these regulations one must look not just at the statute and its legislative history (and, according to the Supreme Court in Brown & Williamson, any number of other sources), but also at the regulations themselves.  The reflexive conclusion that anything the IRS does under Section 385 is blessed is too dismissive of too important a question.  Yes, the authority granted under Section 385 is broad, but it is certainly not limitless.

In October 2016, the IRS issued final regulations (the “Final and Temporary 385 Regulations”, and together with the regulations proposed on April 4, 2016, the “Regulatory Package”),[9] trimming some, but not all, of the excesses of the Proposed 385 Regulations.  However, the IRS did not back down from its assertion of regulatory authority.[10]

In this article, we consider whether certain provisions of the Regulatory Package (including regulations proposed and issued under Section 385, Section 956, Section 7701(l) and Section 7874) are outside the scope of the IRS’s authority.  We approach these questions first by formulating our own set of “underlying principles” that we believe are necessary guideposts to any question of regulatory authority.  We believe using our own “underlying principles” as the lens through which to focus the question allows us to develop tax-specific examples to illustrate the limits of regulatory authority rather than rely on the facts of non-tax cases in which similar questions of regulatory authority were considered.  Not surprisingly, the principles we outline below are echoed in the case law involving questions of statutory interpretation and the Administrative Procedure Act (the “APA”).

Even with the changes to the Regulatory Package, as well as potential tax reform which may further limit the impact of any Section 385 regulations,[11] attempting to define the limits of regulatory authority is as important a task as ever.  The expansion of executive power, the seeming impossibility of bipartisan legislation, and the potential of radical tax reform which would involve, inevitably, Congress charging the Treasury Department and the IRS to fill in statutory gaps mean the limits of administrative authority will continue to be tested in the near future.


Below we lay out principles necessary for discerning the limits of regulatory authority.  These principles are not intended to be comprehensive or even the most fundamental principles of the limits of regulatory authority (indeed, in proving out our propositions, we rely on more fundamental concepts of law).  We chose to highlight the below because each is integral in examining whether specific provisions of the Regulatory Package exceed IRS authority.

A.     A grant of regulatory authority needs to be read precisely in order to determine whether an interpretation of that authority is unreasonable.

Of course it does.[12]  Yes, everyone understands that the executive has been given wide latitude to craft reasonable interpretations of the questions that have been granted to it.[13]  That probably also means (although it is a different question) that the executive has been given wide latitude to craft reasonable interpretations of the scope of the questions.[14]  And, in many cases, there will be many reasonable interpretations of the scope of those questions.  However, it does not follow that there are infinitely many; some will be unreasonable.  And we can only discover the answer to that question by really digging into the initial grant of authority by the legislature.

Let’s illustrate by a real example (which will be discussed later)—Section 956(e).  Section 956(e) authorizes “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.”  That sounds very broad, and it is, but not infinitely so.  There are many textual uncertainties, and one’s view on the scope of authority granted by Section 956(e) will depend on how one interprets those uncertainties.

The first question is what Congress means when it references the “purposes” of Section 956.  One possible answer could be that the purposes of Section 956 are to treat as a deemed dividend the portion of the controlled foreign corporation’s (CFC’s) un-repatriated, non-previously taxed income (PTI) earnings that have been invested in U.S. property.  A second possible answer could be that the purposes of Section 956 are to treat as a deemed dividend the portion of the CFC’s un-repatriated, non-PTI earnings that have been deployed in a manner that produces an economic benefit to the U.S. shareholder that is sufficiently analogous to a dividend that it should be treated as such.  The difference in effect is obvious.  And although it may seem intuitively correct to say that the latter is at least a reasonable interpretation, we are not so sure that is true.  The latter was certainly the impetus behind the legislative effort, and was the subject of the debate. But the outcome of the debate was a significantly narrower phraseology.  Doesn’t that imply a legislative rejection of the cases that fell outside of that phraseology?[15]  Again, suppose the House advanced a broader definition of U.S. property for purposes of Section 956, the Senate rejected it, and the conferees supported the Senate version.  Could the executive reinstate the House version in regulations under the auspices of a broad interpretation of the “purposes” of Section 956?  That seems like a stretch.[16]

The second question involves the use of the word “including.”  In this context, does “including” mean that the second clause is a subset of the first clause, or is in addition to the first clause?  It is incredible that such a basic point could be uncertain, but it seems so.  The import is not as dramatic as the first question, but it is important nonetheless.  If regulations do not further the purpose, does the executive still prevail if the regulations prevent avoidance?  Unclear.

The third question involves the word “provisions.”  Here, the foil is obviously the prior word “purposes.”  “Provisions” certainly reads as a narrower term than “purposes,” more akin to “what I wrote down on the piece of paper” than “the thoughts in my head as I was writing.”  Does that narrowness necessitate a reinterpretation of the breadth of “purposes,” which was so important to the first question?

The fourth question is what to make of the shift from the positive to the negative.  The first clause empowers regulations “to carry out the purposes”; the second clause empowers regulations “to prevent the avoidance of the provisions.”  Perhaps that phrasing implies that one should be restrictive in interpreting the first grant but liberal in interpreting the second?

The fifth question involves the last phrase, “through reorganizations or otherwise.”  That sounds like the legislators had something specific in mind.  If we find something specific in the legislative history (which turns out to be an invocation to prevent arbitraging Section 956 against now-repealed Section 956A),[17] does that mean that we should reject that there is any additional content to this phrase?  And, last but not least, does the invocation of reorganizations mean to lower the otherwise applicable standards, such that what might not have been found permissible in general is suddenly permissible if a reorganization is involved?

These are hard questions, no doubt.  We believe there are many reasonable answers, and the executive should prevail if its regulations can be supported by any of those.  But some answers are unreasonable, and if the executive’s regulations can only be defended on that basis, then they have a problem.

B. A regulation is not authorized if it violates another, separate, principle of law.[18] (And authority to promulgate broad regulations does not demonstrate authority to promulgate every narrower-included subset of such regulations).[19]

Assume, for example, that it is true (which we believe it is) that sufficient authority has been delegated under Section 385 to promulgate regulations that recharacterize all related-party debt as equity.[20]  Does that mean that no further inquiry is necessary if the regulations take an even narrower approach?  If the executive has the power to turn all such instruments into equity, does that not imply that it has the power to turn a subset of such instruments into equity?

Suppose the narrower regulations recharacterized related-party debt only in cases where the CEO of the US company was a woman?  A silly example, perhaps, but an illustrative one.  These regulations would be improper because their narrow enforcement violates a separate and in this case clearly superseding principle of law.[21]  This conclusion holds whether the law being violated is supreme (the Constitution) to the law authorizing the regulations or a law passed by Congress that speaks directly on the matter.[22]

Now let us try a slightly more realistic example.  In Stephen Shay’s 2014 article that planted the seed for the Obama administration’s sense of optimism on expansive regulatory authority, Shay postulated that regulations under Section 385 could recharacterize related-party debt into equity so as to achieve the same amount of interest disallowance as would have been achieved if Section 163(j) had been amended to reduce the 50% adjusted taxable income (ATI) limitation to 25%.[23]  This is quite a bold assertion, because the hypothetical regulation is not a rule about Section 385 but rather a rule about Section 163(j).  If we assume that there is no regulatory authority to alter Section 163(j)—and in fact an implicit but clear legislative demand not to do so (more about that later)—then how is this example different from the one involving female CEOs?  In other words, the selective enforcement of Section 385 rewrites (i.e., violates, or overturns) another law.[24]  If this Section 163(j) change could be made, then presumably the Administration could also have used Section 385 to increase the tax rate under Section 11 from 35% to 45%.[25]  A permissive approach to selective application of the Section 385 regulations would (because interest deductions are nearly ubiquitous) permit revision of virtually every other Code section.  Want to repeal the mortgage interest deduction?  Use Section 385.[26]

Want to double the Section 4985 excise tax by imposing a mirror version on corporations?  Use Section 385.  It is true that eventually you run out of ammunition, but Section 385 is a well-stocked arsenal.

C.     A statute, as drafted, is the complete legislative expression on the matter.

Let us now digress briefly to set forth a theory of legislative completeness.[27]  This is important in figuring out whether a violation of another, separate principle of law has occurred.  The legislative process is not orderly.  Bills are cobbled together to please different constituencies, and theoretical clarity is often sacrificed.  The House has one version; the Senate has another; and the negotiated resolution is not designed to please the theoretician.[28] Nevertheless, one must respect the outcome as the complete expression of the legislature on the topic.[29]  If the executive branch is empowered to go beyond merely curing ambiguities, then what is to prevent the executive from reasserting the House position, for example, replaying the legislative debate, but with a different outcome?  It is true that the legislature sometimes asks the executive branch for help in drafting rules.  But the general baseline is that the statute as finally drafted is complete.  If a statute defines bad behavior and imposes a penalty, in general it is not appropriate for the executive to add an additional penalty.[30]  This is true even if the behavior is deeply offensive and is not deterred by the extant statutory penalty.  This is true even if every member of Congress subsequently decries the inefficacy of the statutory language; if those outcries result in an amendment to the statute, then fine, but otherwise, the prior statute remains the complete legislative expression on the matter.  Using a selective application of the Section 385 regulations to add to a statute that is universally viewed as defective is just as problematic as using the regulations to alter a statute that is universally loved.

In order to make the foregoing more concrete, let us return to Section 163(j).  In the prior Section 163(j) example, we analyzed an attempt to change a 50% limitation to 25% through the auspices of broad Section 385 regulatory authority.  In that case it was clear that the statutory rule was being altered because one needed to cross out the number “50” and replace it with the number “25.”  But is it any different if the new provision adds words without deleting anything?  Imagine, for example, that a new subsection was added to Section 163(j) to provide that if a corporation has “disqualified interest” within the meaning of Section 163(j)(1)(A), then in addition to the foregoing Section 163(j) restrictions, the tax rate on the corporation’s net income from all sources is increased from 35% to 40%.  This could be achieved, for example, by providing that, for any corporation with disqualified interest, the amount of disqualified interest is grossed up to arrive at an effective 40% tax rate on the corporation’s net income.  In effect, this change would be an increase in the Section 163(j) “penalty” for overleverage.  If a selective application of Section 385 regulatory authority were used to back into the same result—using Section 385 to add additional penalties for running afoul of Section 163(j)—such action should be analyzed in the same manner as if the regulations had changed 50% to 25%.

D.     A selective regulatory interpretation is not improper merely because it has an effect on other Code sections, even if those other Code sections are acknowledged to be unambiguous and complete.

This sounds like the opposite of what we set out in Part I.2. above, but it is obviously also true.  For example, a selective interpretation of Section 385 has a direct effect on Section 163.  This is not strange; of course, it is the precise reason that one makes a change to Section 385.  Turning an instrument into equity is designed to invoke all of the consequences that follow therefrom.  In plainer language, it doesn’t make sense to say that one is changing Section 385 in order to make a sub rosa change to Section 163; instead, one is changing Section 385 to produce a derivative consequence under Section 163.B This observation immediately makes clear that it is going to be harder to apply these principles than it originally appeared.  We will need to decide when one statutory provision is derivative of another and when they are independent of each other.

Moreover, the effect on another Code section could be a proper application of selective regulatory interpretation of Section 385, not only because the other Code section derives from Section 385, but also because both Section 385 and the other Code section derive from a common higher-order principle.  Thus, for example, one might say that a selective transformation of related-party debt into equity has a suspicious effect on treaties.  Isn’t this a way to backhandedly renegotiate all treaties that have a lesser rate of withholding on interest than on dividends?  We would argue that those suspicions are ill-founded, because both Section 385 and treaties acknowledge the higher-order principle that relatedness is relevant.  Relatedness is relevant to the debt-equity question because it calls into question whether the parties will act purely in accordance with their debtor-creditor status.  Relatedness is relevant to treaties because the portfolio interest exception obviates the need to rely on reduced treaty withholding rates on interest in most other circumstances.

And, moreover still, when it comes to higher-order principles, some are more equal than others.  Suppose, for example, that the presumed higher-order principle is appetite for risk.  From that derives both aggressive structuring to create interest deductions and the incurrence of the substantial understatement penalty under Section 6662.  Would it then be appropriate to selectively apply the related-party debt recharacterization regulations under Section 385 only to companies that had incurred the substantial understatement penalty within the last ten years?  It seems unfair—a continuing cloud of potential recidivism hanging over a taxpayer that has paid full penance for its sins—and unwieldy.  But is it improper?  On balance we would say it is improper, on the theory that the vague correlation to a higher order principle is insufficient to overcome a more direct objection that one has used Section 385 to impose an additional consequence to (i.e., re-write) Section 6662.

E. A selective application of regulatory authority could fail for lack of germaneness, even if other laws/Code sections are not affronted.[31]

To illustrate here, suppose that regulations under Section 385 impose equity treatment on related-party debt where the name of the U.S. company begins with the letter “A”.  This is not gender discrimination; nor does it impinge on any other aspect of the Code.  It is a nonsensical rule.  But is it improper?  In other words, is there a germaneness requirement to regulatory selectivity?

Well, yes, there is.  This is an important observation which is very different from the foregoing ones.  The foregoing principles tried to discern whether the executive was pretending to apply Section 385 but actually doing something else; there was a sense of trickery in the air.  Here the executive is doing nothing nefarious, rather simply something irrelevant.

This is very much at the heart of the matter when we get to analyzing the actual provisions of the Final and Temporary 385 Regulations.  Is the selectivity of applying the debt-equity rules only to debt that is created in a certain manner a relevant distinction, or an irrelevant one?  Is it more like prioritizing relatedness as a super factor (legitimate) or is it more like demonizing the letter “A” (illegitimate).

While the executive is entitled to deference in this context, there must be limits.  It will be easiest to explore this proposition further when examining the actual regulations.


We turn now to four specific provisions of the Regulatory package.  For each provision, we first briefly describe the mechanics of the regulation and the authority pursuant to which the IRS claims it relied.  We then look more closely at the IRS’s claimed authority, and finally ask the question of whether the regulation is within or outside the limits of regulatory authority.

A.     Provision #1: Treasury Regulation 1.956-2T (the “Deemed U.S. Property Rule”)

One of the potential benefits of an inversion transaction is the ability to access “trapped cash” (i.e., cash in a CFC that could not be repatriated to the United States without triggering taxable income, either as a direct dividend or under Section 956).  Following an inversion, a CFC can lend to its foreign parent without triggering a Section 956 inclusion because the loan would not be an “investment in U.S. property” within the meaning of Section 956.  To address this perceived “abuse,” Treasury Regulations Section 1.956-2T(a)(4) provides that, in certain circumstances, stock or an obligation of a non-U.S. person will be treated as U.S. property (we refer to this as the “deemed U.S. property rule”).  More specifically, if an “expatriated foreign subsidiary”[32] holds an obligation or stock of a non-CFC related person that was acquired in relation to the inversion transaction or within the ten years following the inversion transaction, such obligation or stock will be deemed to be U.S. property for purposes of Section 956.

1.    IRS Claimed Authority

According to the Preamble to the Regulatory Package, a transaction in which a CFC that is an expatriated foreign subsidiary lends directly to foreign parent, bypassing its U.S. shareholder(s), is a transaction that circumvents the purpose of Section 956.[33]  The Preamble recites Section 956(e) (described above) and then states, “an inversion transaction may permit the top corporate parent in the newly inverted group, a group still principally comprised of United States shareholders and their CFCs, to avoid section 956 by accessing the untaxed earnings and profits of the CFCs without a current tax to the U.S. shareholders.  This is a result that the U.S. shareholders could not achieve before the inversion.  The ability of the new foreign parent to access deferred CFC earnings and profits would in many cases eliminate the need for the CFCs to pay dividends to the United States shareholders, thereby circumventing the purposes of section 956.”

2.     A Closer Look

As noted above, defining the purpose of Section 956 is not a straightforward exercise.  But before we go there, we must acknowledge this rule for what it is—an anti-inversion rule.  In a foreign-parented structure that includes U.S. corporations and their CFCs, and where the structure is not the consequence of an inversion, the deemed U.S. property rule does not apply.[34]  We can thus infer two separate but related points.  First, this rule is not a rule about Section 956, but a rule about Section 7874.  The deemed U.S. property rule creates another consequence—a penalty—for undertaking an inversion transaction.  Second, the “purpose” of Section 956 is not abused when a non-inverted foreign-parented group with U.S. corporations and lower-tier CFCs undertakes the kind of transaction that “circumvents” the purpose of Section 956 when undertaken by an inverted group.

3.     Is the Deemed U.S. Property Rule within the Scope of the IRS’s Regulatory Authority?

Imagine Section 7874 had never been enacted and an “inversion transaction” was not a technical term but just a colloquial phrase for a tax-free reorganization in which a U.S. target is merged with and into a foreign acquiring corporation.  Could the IRS have issued regulations under Section 956 applying the deemed U.S. property rule only when an “inversion transaction” has previously taken place?  The answer to that question would rest solely on whether such a regulation is within the grant of the authority under Section 956(e).  Since there is no anti-inversion statute in this hypothetical, there is no concern that the deemed U.S. property rule is invalid because it is really a rule about Section 7874.

Section 7874 does exist, however, and so we must ask (in addition to the “purpose” question) whether the deemed U.S. property rule is doing something to Section 7874 that could not be done directly—i.e., is the regulation rewriting Section 7874 in an inappropriate manner?

In effect, the deemed U.S. property rule treats a foreign parent as a U.S. corporation for purposes of Section 956 in the context described above in order to target a perceived abuse relating to inversions.  Had the ownership percentage with respect to the foreign parent’s U.S. shareholders been 80% or more, the foreign parent would have been treated as a U.S. corporation for all purposes of the Code pursuant to Section 7874(b).  Thus, one way to articulate the effect of the deemed U.S. property rule is that for Section 956 purposes, the regulation lowers the 80% ownership percentage threshold in Section 7874(b) to 60%.

Should this be allowed?  Perhaps the answer is yes, in the context of Sections 956 and 7874, insofar as one could argue that the deemed U.S. property rule does not actually cause a foreign parent to be treated as a U.S. corporation; rather, the regulation causes the stock and obligations of a foreign parent (and other non-CFC related persons) to be treated as stock or obligations of a U.S. person.  But that nuance should not be sufficient to bless all regulatory action under one section that frustrates a Congressional bright-line test set forth in another section.[35]

In the extreme, imagine the IRS issued regulations under each Code section, stating that (i) an inversion transaction presents the potential for abuse of such section, and (ii) therefore, in order to prevent such abuse, solely for purposes of such section, a foreign corporation that was involved in an inversion transaction will be treated as a U.S. corporation (or interests in such a foreign corporation will be treated as interests in a U.S. corporation).  In that scenario, the IRS would have overstepped its authority—the net effect of the regulations would be to lower the 80% threshold in Section 7874(b) to 60%—a result at odds with the clear statutory text of Section 7874.  (As discussed below, regulations issued under Section 7874 that reached the same result would be equally invalid.)

Even if the deemed U.S. property rule can survive a challenge that the rule effects an unauthorized sub rosa change to Section 7874, there is still the question of whether the rule can be supported by a coherent description of the purpose of Section 956.  In order for the deemed U.S. property rule to prevent avoidance of the “purpose” of Section 956, we must be able to articulate a purpose of Section 956 that would be circumvented only in the case when the foreign-parented group is the consequence of an inversion transaction, and not an old-and-cold foreign-parented group.

Maybe the avoidance has to do with the fact that an inverted group is still “primarily composed of United States shareholders and their CFCs”, which is the language the Preamble uses to explain the provision.  But if that is relevant to the purpose of Section 956, why is a recently inverted group any different than an old-and-cold foreign-parented group that consists primarily of U.S. companies and their CFCs?

Maybe what the IRS meant in the Preamble was that it is relevant to the purpose of Section 956 that in an inversion transaction the shareholders of the foreign parent are primarily composed of the old shareholders of old U.S. parent.  In other words, the old shareholders of old U.S. parent previously owned a company that would have been taxed under Section 956 if its CFCs had made loans to the old U.S. parent; now the shareholders own a company that, absent the deemed U.S. property rule, would not be taxed if those same lower-tier companies made loans to the new foreign parent.  That would explain the deemed U.S. property rule only applying to inverted groups, but that articulation of Section 956’s purpose is unsupportable—Section 956 was conceived to police the relationship between a CFC and its U.S. shareholders (within the meaning of Section 951(b)).  If a CFC makes a loan to a U.S. person that is not a U.S. shareholder (within the meaning of Section 951(b)) or a person related to the U.S. shareholder, Section 956 is not implicated.[36]

The Preamble gives another suggestion of Section 956’s purpose: “The ability of the new foreign parent to access deferred CFC earnings and profits would in many cases eliminate the need for the CFCs to pay dividends to the U.S. shareholders, thereby circumventing the purposes of section 956.”  This suggests that cash is somehow finding its way into the U.S. shareholder through a subsequent transaction: the U.S. shareholder needs cash, but the inversion transaction eliminates the “need” to acquire the cash by declaring a dividend from the CFC.

If the abuse is really a transaction that occurs after the loan from CFC to foreign parent, i.e., a subsequent transaction where the foreign parent lends or contributes the borrowed cash to a U.S. shareholder, we agree that such a transaction (or transactions) could be abusive of Section 956.  But that is not what is happening in the deemed U.S. property rule, which does not require that cash ever find its way to the U.S. shareholder.  If the cash stays at the foreign parent, how is that different than if a CFC owned by a publicly-traded U.S. corporation lent money to the public shareholders?  If that is the abuse that the deemed U.S. property rule is targeting, it is hard to understand why this is only a circumvention of Section 956 in the context of inverted groups.  Such transactions are equally “abusive” of the purposes of Section 956 in an old-and-cold foreign parented group.[37]

The deemed U.S. property rule is a rule about inversions.  While that does not necessarily mean it cannot be authorized under Section 956 (although there are strong arguments that it cannot), we are not convinced there is a reasonable articulation of the purpose of Section 956 that can support the deemed U.S. property rule and its application only to inverted groups.  If the rule is not in support of the purposes of Section 956 and/or does not prevent the abuse of Section 956, the rule is not authorized under Section 956(e). 

B.     Provision #2: Treasury Regulation Section 1.7874-8T (the “Serial Acquisition Rule”)

The serial acquisition rule in Treasury Regulations Section 1.7874-8T provides that, for purposes of calculating the ownership percentage (i.e., total vote/value of foreign corporation stock held by former shareholders/partners of the domestic corporation/partnership divided by total vote/vale of foreign corporation stock), the value of stock attributable to a “prior domestic entity acquisition” is excluded from the denominator.  A “prior domestic entity acquisition” is any domestic acquisition that occurred during the 36-month look back period beginning on the signing date of the current acquisition being tested.

1.     IRS Claimed Authority

The IRS cites Sections 7874(c)(6) and (g) as its authority to adopt the serial acquisition regulation.

The relevant section of the Preamble declares serial inversion transactions as inconsistent with the purposes of Section 7874 and analogizes the serial acquisition regulation to the stock exclusion rules of Treasury Regulations Sections 1.7874-4T and -7T.[38]  Treasury Regulations Section 1.7874-4T excludes certain stock of a foreign acquiring corporation issued in transactions in anticipation of an acquisition of a domestic entity from the denominator of the ownership fraction that is used to determine the extent to which the inverted corporation is domestically owned.  Treasury Regulations Section 1.7874-7T similarly excludes from that denominator stock of the foreign acquiring corporation that is attributable to certain passive assets.

2.     A Closer Look

Both Sections 7874(c)(6) and (g) give the IRS a broad grant of regulatory authority, and as with Section 956, it is necessary to parse the language closely to determine what is within the scope of that grant.  Section 7874(c)(6) provides:

The Secretary shall prescribe such regulations as may be appropriate to determine whether a corporation is a surrogate foreign corporation, including regulations to treat warrants, options, contracts to acquire stock, convertible debt interests, and other similar interests as stock, and to treat stock as not stock.

According to the IRS, the serial acquisition regulation is authorized because by excluding shares attributable to a prior domestic entity acquisition from the denominator, the regulation is treating “stock” as “not stock.”

Section 7874(g) provides a grant of authority similar to that in Section 956(e): to “provide such regulations as are necessary to carry out this section, including regulations providing for such adjustments to the application of this section as are necessary to prevent the avoidance of the purposes of this section.”[39]   

3.     Is the Serial Acquisition Rule within the Scope of the IRS’s Regulatory Authority?

Let us start with the limits that apply to both Section 7874(c)(6) and Section 7874(g)—limits that apply to all regulatory authority.  We know that a regulation promulgated under either section would be invalid if it violated another law, as in the gender discrimination example.  The serial acquisition rule does not suffer from any defect in that respect.

We also know that a regulation promulgated under either section would be invalid if it rewrote Section 7874—for example, a regulation that crossed out “80” in Section 7874(b) and replaced it with “60.”  This is a harder question, but we believe the serial acquisition rule does not encroach on any of the statutory language of Section 7874.  Many commentators have expressed frustration that the serial acquisition rule applies regardless of whether the previous acquisitions were done pursuant to a plan that includes the later acquisition.[40]  However, the “pursuant to a plan” concepts that are expressed in the statutory language of Section 7874, as well as the existing regulations, are not incompatible with the serial acquisition rule.

Sections 7874(a)(2)(B) and Section 7874(c)(3) refer to acquisitions as part of a plan and presume an acquisition of all of the properties of a domestic corporation over a four-year period to be part of a plan, but those sections are trying to prevent “creeping acquisitions”[41] of a single entity from falling outside of Section 7874.[42]  Treasury Regulations Section 1.7874-2(e) provides that if, pursuant to a plan or a series of related transactions, a foreign corporation undertakes multiple domestic entity acquisitions, the acquisitions are treated as a single acquisition and the entities are treated as a single entity.  The IRS did not rely on any specific authority granted under Section 7874 when promulgating 1.7874-2(e); instead, the IRS believes this regulation “clarified” existing law and that “the operative rule . . . is not a change from current law.”[43]  We read this as applying either the rule in Treasury Regulations Section 1.368-2(h) (treating multiple corporations as a single corporation “if the context so requires”), as was cited in the preamble to the regulations, or, more accurately, acknowledging that under the step transaction doctrine the IRS could aggregate multiple acquisitions made pursuant to a single plan.[44]

The serial acquisition rule is one step further because it applies to any acquisition done within the relevant time period regardless of whether the acquisitions were pursuant to a plan.  That the serial acquisition rule was issued pursuant to Sections 7874(c)(6)’s and 7874(g)’s authority implies that the IRS believes the rule is not a “clarification” of current law, and we agree.  It would be inappropriate to rely on Section 1.368-2(h) or the step transaction doctrine where separate acquisitions are truly unrelated other than temporally.  But that does not suggest that the rule is somehow inconsistent with the statute; it suggests that for the regulation to be authorized, the authorization must come from Section 7874(c)(6) or Section 7874(g).  And so we must to turn to the language of those sections.

Section 7874(c)(6) grants broader discretion, but is narrower in scope, than Section 7874(g).  The former allows the IRS to issue regulations “as may be appropriate”, whereas the latter authorizes regulations “as are necessary.”  “As may be appropriate” gives the executive a lot of leeway—the regulations can be appropriate, but not necessary, and the “as may be” language suggests that the executive has quite a bit of room to decide what is appropriate (although Chevron deference likely grants the executive such power in any event).[45]  But Section 7874(c)(6) regulations are limited to determining whether a corporation is a surrogate foreign corporation, whereas the IRS can issue regulations under Section 7874(g) “to carry out this section [7874].”  Since the regulations determine whether a corporation is a surrogate foreign corporation by treating stock attributable to a previous domestic entity acquisition as “not stock”, which is specifically authorized in Section 7874(c)(6)(B), the regulations come within the narrower scope, but broader discretion of Section 7874(c)(6).

We begin again with an example of a hypothetical regulation that would be unauthorized under Section 7874(c)(6), despite the Section’s broad grant of authority.  Suppose the IRS issued a regulation, the preamble to which stated that in addition to the specific stock exclusion rules set forth in the Section 7874 regulations, the IRS has determined that taxpayers are implementing many other transactions that have the effect of increasing the denominator of the ownership fraction in a manner inconsistent with the purpose of Section 7874. In order to combat these transactions (which are too varied and creative to tailor specific rules in response), the IRS is going to automatically discount the denominator by 25%.  The preamble might further declare that this rule, while not perfectly precise, is adopted as a matter of administrative convenience in light of taxpayers’ attempts at manipulating the ownership fraction.  (Indeed, the IRS has recently offered this justification in a number of contexts, some of which raise similar questions of authority.)   The effect of this regulation, of course, would be to lower the statutory percentage thresholds in Section 7874 (and increase a 60% ownership percentage to 80%).  That regulation would fail because it would be a backhanded way of rewriting Section 7874.

Does the serial acquisition rule similarly rewrite Section 7874 in any manner?  We believe the answer is no.  Congress identified a specific abuse in adding the “pursuant to a plan” language in Section 7874(b)—a taxpayer should not be able to avoid section 7874 by structuring one or more related acquisitions as separate transactions.  That does not mean that Congress viewed multiple transactions that are not pursuant to a plan as per se not in avoidance of the purposes of section 7874.  A foreign corporation could have a plan to acquire smaller U.S. companies, growing in size until the foreign corporation could acquire larger and larger U.S. companies without triggering the relevant ownership thresholds.  This scheme might not fall within Section 7874(b)’s “pursuant to a plan” definition because the foreign corporation’s successive targets may not have been identified at the time of previous transactions. Nevertheless, one might still consider such a fact pattern as inconsistent with the purpose of Section 7874.  The fact that Congress identifies a specific abuse and targets such abuse in the statute does not preclude an agency from regulating other situations the agency finds abusive, particularly where Congress has granted broad anti-abuse rulemaking authority and Congress has not directly commanded otherwise with respect to the relevant situation.

Serial inversions were a likely target of any IRS action and there were a number of ways the IRS could have attacked the issue.  For example, the IRS could have adjusted the numerator of the ownership fraction by declaring that all stock issued in prior domestic entity acquisitions to be “by reason of” stock (essentially aggregating all domestic entity acquisitions in the last three years).  That approach might have been sustained, although it would have been outside of the specific grant of authority under section 7874(c)(6) “to treat stock as not stock”.  Alternatively, the IRS could have issued regulations deeming all domestic entity acquisitions within the last three years as having been adopted pursuant to a plan.  That too might be authorized, although one might argue that a hard-and-fast rule, as opposed to a presumption, might be beyond the statutory text if a taxpayer could prove the absence of a plan.  The IRS instead relied on the authority in Section 7874(c)(6), and we believe it was within its power to do so.

It is worth noting that this approach might become impermissible if taken to extremes.  If the serial acquisition rule were not limited to a three-year look-back, but a twenty- or fifty-year look-back, at some point the rule would become too tangential to the activity the rule was intending to police.  We do not believe that is the case here, although the question of “how long is too long” is undoubtedly a question that cannot be answered by any set of objective guidelines on limiting regulatory authority. We leave you with the unsatisfying conclusion that the serial acquisition rule does not seem like an overreach. 

C.     Provision #3: Treasury Regulation Section 1.7701(l)-4T (the “De-CFC Rule”).

Following an inversion transaction, a foreign parent and its affiliates may undertake transactions to dilute a U.S. shareholder’s interest in an expatriated foreign subsidiary, in some cases causing the expatriated foreign subsidiary to lose its status as a CFC.  These transactions have the potential to allow a CFC’s pre-inversion earnings and profits (E&P) to be distributed to a non-CFC related person (i.e., a foreign parent or one of its foreign subsidiaries) such that the E&P permanently avoids attracting U.S. tax.  In order to address this perceived abuse, Treasury Regulations Section 1.7701(l)-4T (the “de-CFC rule” or “de-CFC regulation”)[46] recharacterizes a “specified transaction” in which an expatriated foreign subsidiary issues stock to a “specified related person” during the applicable period.  The base example of a specified transaction in the de-CFC regulation is set forth in Treasury Regulations Section 1.7701(l)-4T(g) and is as follows:  DT is owned by FP following an inversion transaction; DT owns CFC, and FP also owns FA.  FA subscribes for stock representing 60% of the vote and value of CFC in exchange for $6x of cash.  Absent recharacterization, CFC would cease to be a CFC as a result of FA’s ownership. Under the recharacterized transaction, FA is treated as having contributed $6x of cash to DT in exchange for DT stock.  DT is deemed to contribute $6x of cash to CFC for additional stock of CFC and the terms of the deemed DT stock and deemed CFC stock mirror each other and mirror the terms of the actual CFC stock (issued to FA).     

1.     IRS Claimed Authority.

The Preamble (and Notice 2014-52) state that the IRS is relying on Section 7701(l) for the authority to issue the de-CFC regulation.  The entirety of Section 7701(l) reads: “The Secretary may prescribe regulations recharacterizing any multiple-party financing transaction as a transaction directly among any 2 or more of such parties where the Secretary determines that such recharacterization is appropriate to prevent avoidance of any tax imposed by this title.”

The Notice claims that Section 7701(l)’s legislative history authorizes the IRS to issue regulations under Section 7701(l) to combat abuse under “a wide variety of Code sections.”[47]

2.     A Closer Look.

First, neither Notice 2014-52 nor the Preamble acknowledge the title of Section 7701(l): “Regulations Relating To Conduit Arrangements.”  While the legislative history of Section 7701(l) does not limit the IRS to “back-to-back loans,” it is clear from the title of Section 7701(l), the case cited to in the legislative history, and the general discussion of Section 7701(l) at the time the provision was adopted, that Congress was targeting transactions (i) with more than two parties, and (ii) where certain parties (conduits) were introduced to the transaction in order to avoid U.S. tax that otherwise would have been imposed if the transaction had been undertaken without the conduit’s involvement.

3.     Is the De-CFC Rule within the Scope of the IRS’s Regulatory Authority?

In a word, no.  One does not need to perform a deep dive into the legislative history or a particulate analysis of the statutory text (but we’ll get there) to conclude that the de-CFC rule should not be something the IRS can adopt under the power granted by Section 7701(l). The statute was intended to grant the IRS regulatory authority to prescribe rules to recharacterize transactions involving more than two parties and more than two legal arrangements as more direct transactions involving fewer parties and fewer steps.  The de-CFC rule creates additional steps, additional legal arrangements, and deems parties that were not party to the transaction to have been involved.[48]  The IRS has exercised its authority under a statute designed to disregard conduit arrangements in order to create a deemed conduit arrangement.

In addition to being outside the scope of congressional purpose behind the anti-conduit regulations, the de-CFC rule also exceeds the explicit regulatory grant of Section 7701(l).  Section 7701(l) authorizes the IRS to recharacterize “any multiple-party financing transaction as a transaction directly among any 2 or more of such parties.”  Even if one reads “multiple-party” as encompassing two-party transactions, the IRS is authorized to recharacterize the transaction only as a transaction among two or more of such parties—i.e., the parties to the multiple-party financing transaction.  In the example above, the de-CFC rule recharacterizes a transaction between two parties (FA and CFC) as having occurred between more than two parties (FA, DT and CFC).  Because DT was not a party to the original financing transaction, the anti-conduit regulation cannot recharacterize the transaction as having included DT.  The de-CFC rule is outside of the regulatory authority granted by Section 7701(l), a conclusion supported by a common sense understanding of “conduit” and the clear text of the statute. 

D.     Provision #4: Treasury Regulation Section 1.385-3 (the “Per Se Rule”).

Section 385 authorizes the IRS to prescribe regulations to determine whether an interest in a corporation is stock or debt (or part-stock/part-debt).[49]  The regulations are supposed to set forth factors to be taken into account to determine “with respect to a particular factual situation whether a debtor-creditor relationship exists or a corporation-shareholder relationship exists.”[50]

In Treasury Regulations Section 1.385-3, the IRS set forth a “per se rule” that treats debt as equity in the following circumstances where the debtor and creditor are members of the same expanded group:[51] (1) the debt is distributed from a corporation to its shareholder, (2) the debt is exchanged for expanded group stock (subject to certain exceptions), or (3) the debt is exchanged for property in an asset reorganization.  In addition, if a corporation borrows cash from an expanded group member (a “funding transaction”) with a principal purpose of engaging in a transaction described above, but where cash or other property is distributed or exchanged instead of debt (a “funded transaction”), the funding transaction is recharacterized as equity (the “funding rule”).  The funding rule is implicated if the funding transaction and funded transaction occur within 36 months of each other, regardless of whether the funding transaction was undertaken with a principal purpose of financing the funded transaction.[52]  There are exceptions to the per se rule, including (1) an exception to the extent of accumulated E&P of the debtor,[53] (2) a de minimis threshold of $50 million,[54] (3) an exception for debt between consolidated group members,[55] (4) a limited ordinary course of business exception,[56] (5) an exception for instruments issued by certain regulated financial companies and regulated insurance companies,[57] (6) an exception for debt issued by non-U.S. corporations, and (7) a rule allowing certain distributions and acquisitions by an expanded group member to be netted against the fair market value of stock issued by that member.[58]  If a debt instrument is recharacterized as stock, the debt instrument is treated as stock for all purposes of the Code.

1.     IRS Claimed Authority.

In the Preamble to the Proposed 385 Regulations,[59] the IRS relied on Section 385(a)’s grant of regulatory authority to adopt rules for the “particular factual situations” that the IRS believes “raise significant policy concerns.”[60]  The IRS cited to Section 385’s legislative history for the proposition that the IRS need not rely on the factors listed in Section 385(b):

The provision also specifies certain factors which may be taken into account in these [regulatory] guidelines. It is not intended that only these factors be included in the guidelines or that, with respect to a particular situation, any of these factors must be included in the guidelines, or that any of the factors which are included by statute must necessarily be given any more weight than other factors added by regulations.[61]

The Preamble cited debt/equity factors that were considered in cases involving related parties.[62]  More specifically, in adopting the per se rule in the context of distributions of debt instruments, the IRS determined that “the parent-subsidiary relationship, the fact that no new capital is introduced in connection with a distribution of debentures, and the typical lack of a substantial non-tax business purpose, support the conclusion that the issuance of a debt instrument in a distribution is a transaction that frequently has minimal or nonexistent non-tax effects.”[63]  The extension of the per se rule to exchanges and acquisitions involving related-party debt, as well as the funding rule, were adopted because the IRS believes such transactions raise similar concerns.

2.     A Closer Look.

Section 385’s grant of regulatory authority is indeed very broad.  While the statute lists certain debt/equity factors, the legislative history is clear that the IRS is free to issue regulations that consider other factors and/or do not consider the listed factors at all, and the IRS has a wide discretion with respect to the weight given to these factors in particular factual situations.[64]

3.     Is the Per Se Rule within the Scope of the IRS’s Regulatory Authority?

Before we address whether there are any limitations on the IRS’s authority under Section 385 (there are) and whether the IRS has exceeded its authority in adopting the per se rule (we believe the IRS has), it is necessary to consider what Congress thought it was authorizing when it enacted Section 385.

Outside the context of the Section 385 Regulations, the question of whether an instrument is debt or equity is, and has always been, a question of economic substance: is the relationship between the parties that of a debtor-creditor or that of a corporation-shareholder, and have the parties acted in a manner consistent with that relationship?  The factors referenced in Section 385, the debt/equity case law, and IRS public and private rulings are guideposts to assist in determining the true economic relationship between the parties, which is the essence of the debt/equity question.  The legislative history of Section 385 (during its enactment and subsequent amendments) evidences that Congress shares this view.  For example, the House Report accompanying the 1989 amendment to Section 385 stated “The characterization of an investment in a corporation as debt or equity for Federal income tax purposes generally is determined by reference to numerous factors that are deemed to reflect the economic substance of the investor’s interest in the corporation[65] (emphasis added).  Three years later, when Congress again amended Section 385, the House Report discussing the amendment provided:

In 1969, Congress granted the Secretary of the Treasury the authority to prescribe such regulations as may be necessary or appropriate to determine whether an interest in a corporation is to be treated as stock or indebtedness for Federal income tax purposes (sec. 385).  The regulations were to prescribe factors to be taken into account in determining, with respect to particular factual situations, whether a debtor-creditor relationship or a corporation-shareholder relationship existed.[66]

Similarly, the IRS acknowledged in the Preamble to the Proposed 385 Regulations that the debt/equity question is a question of economic substance.  In fact, the entire premise of the documentation requirements set forth in Treasury Regulations Section 1.385-2 is that (i) there are certain factors (e.g., ability to repay) that indicate whether as a matter of substance an instrument is debt or equity, (ii) in the related-party context, it is less likely that a creditor will have demanded evidence of those factors, and (iii) therefore, it is necessary to impose documentation requirements in order to ensure the IRS has the necessary information in order to undertake a substantive analysis of the instrument in question to determine whether it is debt or equity.[67]

In contrast, the per se rule is fully divorced from the question of whether an instrument is, in substance, debt or equity.[68]  In fact, the per se rule first requires a taxpayer go through a traditional debt/equity analysis (i.e., determine the economic substance of the instrument).[69]  Only if such an analysis reveals the instrument as true debt does the per se rule potentially recharacterize the debt (which was just determined to be debt under the traditional debt/equity factors) as equity.  We think it is fair to say that this is not what Congress envisioned when it adopted and subsequently amended Section 385.

The question of what is authorized under Section 385 is admittedly a harder question, given the expansive authority granted by the statutory language and the legislative history.  Lee Sheppard has written that “the statute clearly empowers the government to write regulations that would effectively replace case law.”[70]  That may be the case in some respects, but it cannot be true in all respects.  Suppose the IRS issued a regulation under section 385 that declared all instruments with a term of less than one year, regardless of their relationship to the issuer/holder, to be equity.  Yes, the term of an instrument can be relevant to the debt/equity analysis, but surely a shorter-term instrument is more debt-like, not more equity-like.  Would this rule be authorized?  The answer must be no.  In laypersons’ terms; that would be a crazy rule. In the language of a judicial challenge of administrative action; such a rule could never survive State Farm.[71]  Common sense tells us that this rule cannot be within the IRS’s scope of authority, and  APA jurisprudence tells us the same.[72]

What is the thing that makes such a rule unauthorized?  The answer is that the rule bears not just zero relation, but an inverse relation, to the substance of the instrument as debt or equity.  That’s what makes it a crazy rule; that’s what would make its issuance “arbitrary and capricious”[73]. As the Supreme Court recently noted, “Even under Chevron’s deferential framework, agencies must operate ‘within the bounds of reasonable interpretation.’”[74]  Such reasonable interpretation “must account for both ‘the specific context in which . . . language is used’ and ‘the broader context of the statute as a whole.’”[75]] The example above, however absurd, proves that we have to look at the per se rule and the relationship of the rule to the substantive question of debt vs. equity to see whether the rule is authorized.  The broad authority granted to the IRS in Section 385 – and it is broad – has limits.  And, more importantly for the analysis of the per se rule, the actual substance of an instrument as debt or equity, and the relation any proposed rule has to that substance, is one of those limits.

The Preamble describes or cites to a few debt/equity cases, focusing primarily on Kraft v. Commissioner.[76]  In Kraft, the taxpayer was a subsidiary corporation that had issued a debt instrument to its parent and claimed deductions for interest payments made with respect to the instrument.[77]  The court held for the taxpayer, stating that “to strike down a genuine transaction because of the parent-subsidiary relation would violate the scheme of the statute [permitting deduction of interest payments] and depart from the rules of law heretofore governing intercompany transactions.”[78]  The Preamble, however, states that “the factors discussed in Kraft and Talbot Mills, including [1] the parent-subsidiary relationship, [2] the fact that no new capital is introduced in connection with a distribution, and [3] the typical lack of substantial non-tax business purpose, support the conclusion that the issuance of a debt instrument in a distribution is a transaction that frequently has minimal or nonexistent non-tax effects.”

We believe the first factor is (or can be) relevant to the determination of the economic substance of an instrument.  In the context of proportionately-held debt (e.g., parent-subsidiary debt), there is limited economic significance to the distinction between the debtor-creditor and corporation-shareholder relationships.  Moreover, the parties in such an arrangement may be less likely to act in accordance with the form of their legal arrangements.  If debt and stock are held in the same proportion and there are no third-party creditors, why would the creditors (who are also shareholders) enforce their legal rights in an event of default?

In addition, this factor has universally been considered as relevant in the debt/equity case law and in IRS guidance.[79]  It is true that Section 385 and its legislative history make clear that case law does not bind the IRS in considering or weighting factors in issuing regulations under Section 385.  But courts and the IRS have not come upon this factor by accident – this factor is well-established as a relevant data point because the overlap of the creditors and shareholders of a corporation may call into question whether the purported debt instrument creates a true debtor/creditor relationship and whether the parties will act in a manner consistent with that relationship.  In other words, the case law confirms the conclusion that this factor is germane to the substance of an instrument as debt or equity.  This is in contrast to the “no new capital factor” discussed below, which, but for the few cases cited in the Preamble to the Proposed 385 Regulations, is absent from the debt/equity case law.

It is worth discussing further why the related party factor is germane to the debt/equity question.  Debt issued to a shareholder or related party does not, per se, make the debt “equity-like” in the same way other traditional equity-like features play into the analysis.  Convertibility into equity, for example, is an equity-like feature regardless of context (i.e., regardless of to whom and in what context the debt is issued).  So too are features like subordination, the lack of a fixed maturity, the lack of interest, voting rights, and so forth.  The related party factor is relevant for a different reason—because the relationship between the parties calls into question whether a true debtor/creditor relationship exists, regardless of the legal rights that the instrument creates.  It is a situation that raises the antennae of the court (or the tax advisor, or the IRS) because the related party context allows for a greater possibility that the parties will not act in accordance with the terms of their agreement, which is unlikely when debt is issued to a true third party.[80]

When the IRS issued regulations under Section 385 in 1981, the preamble acknowledged that the IRS considered treating all proportionately-held debt as stock, implying that the IRS believed it had authority to do so.[81]  Does the related party context so completely cloud the lens through which a debt/equity analysis is viewed that the IRS is within its authority under Section 385 to recharacterize all related party debt as equity?  We are not sure—there are plenty of other related party contracts that give rise to questions about whether the parties will act in accordance with the contract’s terms, and absent evidence the parties plan to ignore (or have ignored) their legal rights and obligations, such contracts generally are given effect for tax purposes.[82]  In any event, such an approach would have been, in our opinion, more consistent with the IRS’s authority than the approach the IRS adopted with the per se rule.[83]

The “no new capital factor” is not relevant to the question of whether an instrument is in substance debt or equity.  There is no logical connection between the economic substance of an instrument (and the nature of the relationship it creates) and the fact that it was issued as part of a recapitalization or as a dividend, and the vast majority of the case law and IRS guidance has not considered this a relevant factor.  Of the three cases the IRS adduced to support its position (Kraft, Talbot Mills, and Sayles Fishing), the case that considers the issue most thoroughly, Kraft, determined that the fact that the notes were issued in a distribution was not a relevant factor, let alone a determinative one.  The court stated that the Code “recognizes that indebtedness may be created by a distribution or by a recapitalization exchange through the issuance of securities out of capital or earnings or both, though the tax treatment may vary with the method chosen.”[84]

In addition, Congress did not consider the “no new capital factor” to be germane to the debt/equity inquiry in 1984 when Congress added Section 1275(a)(4) to the Code and amended Section 312(a)(4), each of which provide rules that apply when a corporation distributes its own note as a distribution under Section 301.  Neither the Code sections nor the legislative history suggest any special debt/equity scrutiny is appropriate in such context.

Does it matter that the IRS can point to three cases that mention this factor (one of which ignores it) in the more than one hundred cases that have considered the debt/equity question?  If the IRS is not bound by case law, the fact that the case law has or has not considered a particular factor should not bear directly on the question of authority.  If no case had ever mentioned this as a factor, however, declaring such a factor as germane to the debt/equity question would be suspect.  Do three cases move the needle?  We think not.

One does not need to mine the case law to come to the conclusion the “no new capital” rule is irrelevant to the substance of an instrument as debt or equity.  Suppose Corporation X funds its subsidiary in year 1 with $90 of equity, and $10 of debt.  The “no new capital rule” is not implicated, and the question of whether that $10 of debt is true debt will depend on the outcome of an economic substance inquiry based on the traditional debt/equity factors.  Now suppose in year 3 the value of the subsidiary has increased to $200, and Corporation X determines that its subsidiary is over-capitalized relative to Corporation X’s other subsidiaries, and causes the subsidiary to distribute a second $10 note.  Is the relationship between Corporation X and its subsidiary any more debtor/creditor-like with respect to the first note than their relationship with respect to the second note?  The answer will depend on the actual relationship between the parties, the terms of the notes, and so forth.  What it will not depend on is whether the note was issued under the circumstances described in year 1 or those described in year 3.  Another example:  suppose Corporation X and Corporation Y, unrelated, is each formed on day 1.  Corporation X’s shareholders hired counsel prior to formation, and on counsel’s advice, Corporation X was part equity- and part-debt capitalized: $100 contributed in exchange for $90 of stock and a $10 note.  Corporation Y’s shareholders contributed $100 in exchange for $100 of stock, but on day 2, after consulting with counsel, caused Corporation Y to distribute a $10 note.  In each case, introducing debt to the capital structure was tax motivated.  Any reason Corporation Y’s note is more equity-like that Corporation X’s note?  Of course not.[85]

This brings us to the third “factor.”  The third “factor” that the IRS considered is that distributions of debt typically lack a substantial non-tax business purpose.  This is not a factor but rather support for why the no new capital factor is relevant.  Framed differently, because debt distributions are generally tax motivated (a conclusion for which the IRS does not provide any support), debt distributions require special scrutiny from a debt/equity perspective.

This is, simply, a fundamental misunderstanding of the debt/equity question.  Debt is a tax-preferred form of raising capital.  That is how the Code works.  A taxpayer may always choose to issue debt instead of equity, even if that decision is entirely tax motivated, as long as the instrument is true debt.  Suppose a corporation needs to raise capital and prepares to access the public capital markets.  The tax director is on vacation, and no one bothers to consult with her, and the decision is made to raise equity capital.  Moments before the offering documents are to be filed, the tax director rushes in and asks the CFO and COO to consider issuing debt instead of equity because of the tax benefits.  The CFO and COO conclude that the principal and interest payments are supportable by the corporation’s operations, and decide that the benefit of the tax deduction associated with the interest expense exceeds the burdens of the required interest and principal payments.  The corporation instead issues debt securities to the public.  Is any of this relevant to the debt/equity question?  No.  An IRS auditor could have been in the room while these discussions were taking place and the fact that the switch from equity to debt was 100% tax motivated does not color the debt/equity analysis one bit.  If the instrument is in substance debt, and the parties behave in a manner consistent with that substance, the instrument is debt.

The fact that a taxpayer’s motivation is irrelevant to the debt/equity question is not limited to the debt/equity context; it is the case for much of the Code.  A taxpayer is able to structure into a like-kind exchange solely to achieve a specific tax result, or change the mix of consideration in a merger solely to achieve tax-free status.  National Starch and Granite Trust transactions are both, by definition, tax motivated transactions.  The fact that a debt distribution may be tax motivated is no more relevant to the debt/equity question than the fact that a corporation issuing third party debt may have been motivated by the fact that interest is deductible.  Debt/equity is not a context where motivation matters.  The third “factor” also fails the germaneness inquiry.

It is not too surprising that the per se rule is unhinged from the question of whether an instrument is in substance debt or equity, because the IRS acknowledges in the Preamble to the Proposed 385 Regulations that this rule serves a different purpose.  According to the IRS, the Section 385 Regulations are intended to “limit the benefits of post-inversion tax avoidance transactions” (i.e., the Section 385 Regulations are the “earnings stripping” guidance referred to in the 2014 Notice and Notice 2015-79).[86]  The IRS considers a note distribution to a foreign parent from an inverted U.S. corporation to be this type of transaction.

Additionally, in the Preamble to the Proposed 385 Regulations, the IRS considered the Proposed 385 Regulations to combat the “abuse” of a first-tier CFC’s distribution of a note in a year when it has no E&P, which allows for future untaxed E&P to be brought into the U.S. without tax as principal payments are made on the note.[87]

We previously discussed and analyzed an overt use of regulations under Section 385 to adopt an earnings stripping policy inconsistent with the statutory text of Section 163(j)—a set of regulations that recharacterize debt as equity to arrive at a different percentage for calculating “excess interest expense” (25% instead of 50%) in Section 163(j).  Similarly, we asked whether if that approach is impermissible (and it should be), should it be permissible to use Section 385 to rewrite Section 163(j) in a less obvious manner?  Again, we concluded the answer should be no.  Can the policy behind Section 163(j) ever justify regulations under Section 385?  The answer here is less straightforward.  The policies of the two sections are less overlapping than they might seem, as the legislative history of Section 385 (and the inclusion of the Section 385(b) listed factors) suggests that the purpose of Section 385 is not to police interest expense but to provide clarity to the “ambiguities and uncertainties which have long existed in our tax law in distinguishing between a debt interest and an equity interest in a corporation.”[88]   The policy of Section 163(j) (to limit excess interest expense) and Section 385 (to provide clear rules on debt/equity) might be viewed as in service of a similar higher principle, but the further the Section 385 Regulations drift from a substantive analysis of an instrument as debt or equity, the more suspect they become.  The per se rule is completely unmoored from the substantive debt/equity question and, as the IRS acknowledges, motivated entirely by policy concerns.  If this is allowable, the IRS could also issue a “per se rule” for debt of a U.S. corporation held by a non-U.S. person (repealing portfolio interest) because the executive branch determines that the policy of Section 871(h)(3)—the limitations on portfolio interest—needs to be backstopped.

Using Section 385 to support the policy of limiting CFC note distributions in the context described in the Proposed 385 Regulations is even more novel.  When has Congress ever espoused such a policy?  It cannot be found in Section 301, 316, 317, 367(b), 951, or 956.  Indeed, such a policy is arguably inconsistent with those sections.

If the executive branch can use a grant of authority pursuant to Section 385 to backstop a “policy” that is missing, not just from Section 385, but the entire Code, then Section 385’s scope is alarmingly broad—what is to prevent a future administration that determines that the corporate tax is anti-competitive from issuing new regulations under Section 385 that treat all corporate equity as debt?[89]

For any set of Section 385 regulations to be valid, there must be a relationship between the factors the regulations set out and the true nature of an instrument as stock or debt.  The factors listed in the Preamble to the Proposed 385 Regulations do not satisfy this requirement and the per se rule, a rule that the IRS acknowledges is a rule about earnings stripping, is therefore beyond the scope of even the broad authority granted by Section 385.

E.         Provision #5:  Treasury Regulation Section 1.385-3T(f) (the “Partnership Rule”).

Under Treasury Regulations Section 1.385-3T(f), for purposes of applying the per se rule, a controlled partnership is treated as an aggregate of its partners.  Thus, if a controlled partnership borrows, the borrowing is treated as having been made by its corporate partners.  If the debt of the partnership (deemed to be debt of the corporate partners) is recharacterized as equity with respect to one or more of the corporate partners, the holders of the recharacterized debt instrument is deemed to transfer the recharacterized portion of the debt instrument to the corporate partner(s) in exchange for stock of the corporate partner(s).

1.     IRS Claimed Authority.

The IRS breaks down its authorization for the partnership rule into three steps.  First, the partnership rule treats partnership debt as debt of the corporate partners, relying on the legislative history of subchapter K that for purposes of interpreting Code provisions outside of subchapter K, a partnership may be treated as either an entity separate from its partners or an aggregate of its partners, “depending on which characterization is more appropriate to carry out the purpose of the particular section under consideration”.[90]  Second, the IRS analyzes the deemed partner-level debt under the section 385 regulations.  No specific authority is cited for this position, although it follows from the treatment of the partnership debt as partner-level debt under the aggregate theory.  Last, if the deemed partner-level debt would be recharacterized as equity, the IRS relies on the anti-conduit rule of Section 7701(l) to deem the recharacterized debt instrument as having been transferred to a corporate partner in exchange for its stock.[91] 

2.     A Closer Look.

Section 385 is titled “Treatment of Certain Interests in Corporations as Stock or Indebtedness”.  There is no reference to partnerships in any of the legislative history of section 385.  The previous iterations of the section 385 regulations contained no reference to partnerships.  However, partnerships are treated as aggregates of their partners in various provisions of the Code. The text and the purpose behind the anti-conduit rule are discussed in detail above. 

3.     Is the Partnership Rule within the Scope of the IRS’s Regulatory Authority?

Perhaps more interesting than the question of whether the partnership rule is authorized (which we address below) is both Preambles’ discussion of the partnership rule, which informs us of what the IRS believes it can—and cannot—do under Section 385 and Section 7701(l).

The Preamble to the Proposed 385 stated that the IRS does have the authority to recharacterize partnership debt as equity, but that the IRS did not propose that approach because the partnership equity could have given taxpayers the same benefits of interest deductions.[92]  The relevant language in the Preamble read:

If a debt instrument issued by a controlled partnership were to be recharacterized as equity in the controlled partnership, the resulting equity could give rise to guaranteed payments that may be deductible or gross income allocations to partners that would reduce the taxable income of the other partners that did not receive such allocations. Therefore, under the authority of section 7701(l) to recharacterize multiple-party financing transactions, proposed §1.385-3(d)(5)(ii) provides that, when a debt instrument issued by a partnership is recharacterized, in whole or in part, under proposed §1.385-3, the holder of the recharacterized debt instrument is treated as holding stock in the expanded group partner or partners rather than as holding a partnership interest in the controlled partnership.

Under the Proposed 385 Regulations, first, on the aggregate theory, partnership debt would have been treated as partner debt.  Then, partner debt would be analyzed under the Section 385 Regulations to determine whether the debt should be recharacterized.  If it would be recharacterized, the debt would be recharacterized as equity of the partnership under the authority granted pursuant to Section 385.  Then, the partnership equity would be re-recharacterized as equity of the partners pursuant to the anti-conduit rule of Section 7701(l).

The Final and Temporary 385 Regulations take a slightly different approach.  The first step is the same:  a partnership is treated as an aggregate of its partners.[93]  The second step is also the same – the partner-level debt is analyzed under the Section 385 regulations.  The third step outlined in the Proposed 385 Regulations  – recharacterizing partnership debt as partnership equity – is omitted:  “in response to comments, the final and temporary regulations do not recharacterize debt issued by a partnership as equity under section 385.”[94]  Instead, the Final and Temporary 385 Regulations skip right to the final step:

Instead, pursuant to the authority granted under section 7701(l) to recharacterize certain multi-party financing transactions, the temporary regulations deem the holder of a debt instrument issued by a partnership that otherwise would be subject to recharacterization (based on an application of the factors in §1.385-3 to the expanded group partners under the aggregate approach) as having transferred the debt instrument to the expanded group partner or partners in exchange for stock in the expanded group partner or partners.[95]

The change suggests that the IRS acknowledges at least some limit on their power under Section 385.  Unfortunately, the IRS believes it can simply look to the anti-conduit regulations to get to the answer it desires.  As we discuss below, this type of “connect-the-dots” authority—mining for whatever textual support (whether it be in legislative history, case law, or the Code) can support an end result, irrespective if those authorities are being completely severed from their original context, is not appropriate.  But before we get there, let us examine the partnership rule in more detail.

The first question raised by the partnership rule is whether it is appropriate to consider partnership debt as debt of its partners. The treatment of a partnership as an aggregate of its partners or as an entity varies throughout the Code.  For purposes of the portfolio interest 10% ownership test, for example, ownership is tested at the partner level.[96]  The legislative history of the 1954 Act, however, requires that the aggregate or entity theory be adopted based on whichever theory better carries out the purposes of the relevant Code section.  This question too, then, requires us to consider the purpose of Section 385.  If the purpose of Section 385 is to prevent a situation the IRS considers abusive, then the inquiry is over—whatever theory (aggregate or entity) the IRS chooses will necessarily be in furtherance of its anti-abuse provision.[97]  If however, as we have demonstrated above, the purpose of Section 385 is to set forth factors to determine the true nature of the relationship between the debtor and creditor, then the IRS has again strayed from traditional debt/equity analysis in applying the aggregate theory.  Debt/equity authorities look to the relationship of the legal borrower and legal lender, unless there is reason to doubt that the legal borrower is truly the source of the funds on which the lender is relying to repay the debt (for example, in a Plantation Patterns-like fact pattern).  To look through partnerships as a general matter makes no sense in the debt/equity framework,[98] no more so when considering related-party debt.

Once the partnership debt is deemed to be partner-level debt, applying the Section 385 tests to the partner-level debt, particularly as it relates to the funding rule, can lead to odd results.  Suppose a controlled partnership (P) has two partners Parent and Sub1.  P borrows from a member of the expanded group and uses the proceeds in P’s trade or business.  P never makes a distribution or loan to either Parent or Sub1, but within three years of P’s borrowing Sub1 makes a distribution (out of its own funds) to Parent, and none of the exceptions to the per se rule are available.  Because P is treated as an aggregate of its partners, Parent and Sub1 would each be treated as having borrowed from the expanded group member and, because Sub1 paid a dividend, a portion of that borrowing would be recharacterized as equity in Sub1.  The Preamble to the Proposed 385 Regulations states the reason the funding rule is appropriate is “because money is fungible and because it is difficult for the IRS to establish the principal purposes of internal transactions.”[99]  But money is not that fungible; it cannot make its way from P to Sub1 without having been distributed or loaned to Sub1.  Thus, even if one adopts the internal logic of the Section 385 Regulations (that the “no new capital” factor is germane to the debt/equity question and that the funding rule is necessary because cash is fungible), that logic cannot support the partnership rule.

Finally, the IRS relies on the anti-conduit rule to complete its recharacterization of partnership debt to corporate stock, in the case of the Final and Temporary 385 Regulations, by deeming the debt holder to have transferred a portion of the debt to the corporate partner(s) in exchange for its/their stock.  We have discussed the limitations of the conduit rule above, and need not revisit them here except to say that the same analysis applies – the anti-conduit rule is not (and was not intended to be) an unlimited anti-abuse rule.[100]

The partnership rule requires the IRS to jump through three hoops, and for each the rule relies on distinct authority.  All three of the authorities are being used in a questionable manner, and we believe than none of them can justify the uses to which the partnership rule puts them.

The partnership rule, perhaps more than any other Questionable Provision (though the de-CFC rule comes close) lays bare just how “results oriented” the rules set forth in the Regulatory Package are.  When the IRS encountered a perceived roadblock under Section 385 (i.e., a potential limit to its authority), the IRS simply invoked the anti-conduit rule to circuitously arrive at the desired result.  This, in our view, cannot be within the bounds of regulatory authority.  The limits of regulatory authority should not be constrained only by whether drafters can piece together a coherent-enough story through unrelated Code sections, cherry-picked case law and scraps of legislative history to support a policy-driven regulation unrelated to the policy of the Code section under which the regulations are issued.  If that is indeed a valid exercise of executive power, it is hard to put any limit on IRS authority.


Our system of government discourages quick action by the legislative branch, and we are experiencing a political moment of greater than normal legislative paralysis.  The executive branch (and the citizens) may become justifiably frustrated by this inability of the legislature to promulgate much needed laws.  It will become very tempting for the executive branch to fill this void with an expansive assertion of regulatory authority.  As a political matter, we as citizens may need to confront the possibility that stasis is the appropriate outcome of the divisions in the country.  Our goal in this paper is more limited—to try to construct a framework to evaluate whether the assertion of regulatory authority fails as a technical matter.  Careful reading of the enabling statute is a necessary starting point for this inquiry, but the analysis must also encompass whether there are improper hidden goals of the regulation, and whether the regulation is sufficiently germane to its purpose.  If broad agreement can be established regarding the limits of regulatory authority, then we will be closer to fulfilling the proper delineation of the functions of our branches of government.


[1] FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120 (2000).

[2] 21 U.S.C. §§ 321(g)–(h), 393.

[3] 529 U.S. at 163.

[4] Id. at 142.

[5] Unless otherwise specified, all section references are to the Internal Revenue Code of 1986, as amended.

[6] See, e.g., Comment Letter on Proposed Rulemaking Treatment of Certain Interests in Corporations as Stock or Indebtedness, Org. for Int’l Investment (July 7, 2016), [] (“The purpose of section 385 was to provide more general guidance outside of the acquisition context on the difference between debt and equity.  There is no indication that Congress believed that debt incurred to acquire stock of a related or unrelated company to which section 279 did not apply should somehow be suspect under section 385.”); Comment Letter on Proposed Rulemaking Treatment of Certain Interests in Corporations as Stock or Indebtedness, KPMG (July 7, 2016), [] (“We believe the proposed regulations exceed the regulatory authority granted by section 385 and go beyond Congressional intent.  They do not seek to delineate a general debt-equity distinction or to clarify or rationalize debt-equity analysis.  Instead, the proposed regulations would simply treat certain arrangements as equity in order to ensure that interest expense deductions are not available in the case of certain transactions.”); Comments re: Proposed Treasury Regulations under Section 385, Pricewaterhouse Coopers (Apr. 7, 2016), [] (“The Proposed Regulations incorporate none of the traditional factors used by the courts in distinguishing debt from equity: they do not look to the terms of an instrument; they do not look to the strength of an issuer’s cash flows or balance sheet; and they do not take into account the intent of the debtor and creditor in making an advance.  This is not what section 385 provides nor what Congress intended in enacting section 385.”).

[7] Cf. Jasper L. Cummings, Jr., Rite Aiding the Debt-Equity Regulations, 152 Tax Notes 377 (2016) (outlining arguments that might be made in opposition to the validity of the Proposed Regulations, including that “[d]ebt in form is not an interest in a corporation; therefore, on its face, [Section 385(a)] does not authorize regulations treating corporate debt as stock” and that because the statute states that “[t]he regulations ‘shall set forth factors’ to be ‘taken into account’ in analyzing particular factual situations and categorizing the relationship between an investor and the corporation[,] the regulations cannot state preclusive rules for broad categories of factual situations based on an extremely limited set of factors (such as an arbitrarily prescribed type of documentation or a time sequence”).

[8] Stephen Shay, Mr. Secretary, Take the Tax Juice Out of Corporate Expatriations, 144 Tax Notes 473 (2014) at 473.

[9] T.D. 9790, 2016-45 I.R.B. 540, 81 FR 72858-72984 (Oct. 21, 2016).  In January 2017, President Trump issued a memorandum instituting a temporary freeze on most federal regulatory activity.  Reince Priebus, Memorandum for the Heads of Executive Departments and Agencies—Regulatory Freeze Pending Review, White House Office of the Press Sec’y (Jan. 20, 2017), [].  It is unclear what, if any, impact the memorandum has had on the Section 385 regulations.

[10] See generally T.D. 9790, 2016-45 I.R.B. 540, 81 FR 72858-72950 (Oct. 21, 2016).

[11] If interest were no longer deductible, the distinction between debt and equity would be considerably less important from a tax perspective.

[12] See United States v. Mead Corp., 533 U.S. 218, 234–35 (2001) (“This Court in Chevron recognized that Congress not only engages in express delegation of specific interpretive authority, but that ‘[s]ometimes the legislative delegation to an agency on a particular question is implicit.’  Congress, that is, may not have expressly delegated authority or responsibility to implement a particular provision or fill a particular gap. Yet it can still be apparent from the agency’s generally conferred authority and other statutory circumstances that Congress would expect the agency to be able to speak with the force of law when it addresses ambiguity in the statute or fills a space in the enacted law, even one about which ‘Congress did not actually have an intent’ as to a particular result.”).

[13] See Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 843–44 (“We have long recognized that considerable weight should be accorded to an executive department’s construction of a statutory scheme it is entrusted to administer, and the principle of deference to administrative interpretations has been consistently followed by this Court whenever decision as to the meaning or reach of a statute has involved reconciling conflicting policies, and a full understanding of the force of the statutory policy in the given situation has depended upon more than ordinary knowledge respecting the matters subjected to agency regulations.”).

[14] Cf. Mead, 533 U.S. at 234–35 (acknowledging that even where a statute precludes Chevron deference to agency interpretations thereof, “an agency’s interpretation may merit some deference whatever its form, given the ‘specialized experience and broader investigations and information’ available to the agency, and given the value of uniformity in its administrative and judicial understandings of what a national law requires”).

[15] See John F. Manning, Why Does Congress Vote on Some Texts But Not Others?, 51 Tulsa L. Rev. 559, 565–71 (2016) (suggesting that the fact that portions of the legislative history of a given statute were not enacted into law, despite legislators’ awareness that legislative history does not have the force of law, suggests that legislators did not believe that the contents of the legislative history could pass).

[16] In fact, the IRS tried to stretch Section 956 in Ludwig v. Commissioner, 68 T.C. 979 (1977).  In Ludwig, the taxpayer, a U.S. person, pledged shares of stock in “Oceanic”, a CFC (whose primary asset was stock in another CFC), in order to secure a loan from a third party.  As part of the loan agreement, the taxpayer entered into a number of negative covenants.  The IRS argued that the pledge of CFC stock was an indirect pledge of its assets, and thus implicated Section 956.  The Tax Court held for the taxpayer, distinguishing between situations in which a taxpayer receives a benefit in respect of the value of a CFC’s assets and when such benefit is covered by Section 956: “That petitioner realized a benefit from owning and pledging his Oceanic stock to secure the bank loans does not mean that Oceanic’s earnings were invested in United States property within the meaning of section 951.  Neither that section nor section 956(c) reaches every benefit derived from the ownership of stock in a controlled foreign corporation.”  Id. at 988.  In 1980, the Treasury Department amended Treasury Regulations Section 1.956-2(c)(2) to provide that a pledge of at least 66 and 2/3rds percent of the voting stock of a CFC is treated as an indirect pledge of the CFC’s assets.  T.D. 7712 (Aug. 6, 1980).

[17] See Conference Committee Report on P.L. 103-66 (“The conference agreement also provides regulatory authority under which the Treasury is instructed to prescribe such regulations as may be necessary to carry out the purposes of Section 956, and to prevent their avoidance.  Within this authority, the conferees anticipate that the Treasury may prescribe regulations that, for example, would prevent taxpayers from taking advantage of the differences between the excess passive asset rules and the rules of Section 956.”).

[18] See, e.g., Nat’l Family Planning and Reproductive Health Ass’n, Inc. v. Gonzales, 468 F.3d 826, 829 (D.C. Cir. 2006) (“[O]f course[,] a valid statute always prevails over a conflicting regulation.”).

[19] See, e.g., Judulang v. Holder, 132 S. Ct. 476, 485 (2011) (explaining that even though the Board of Immigration Appeals “may well have legitimate reasons for limiting [the scope of a statute permitting discretionary relief] in deportation cases . . . it must do so in some rational way.  If the BIA proposed to narrow the class of deportable aliens eligible to seek . . . relief by flipping a coin—heads an alien may apply for relief, tails he may not—we would reverse the policy in an instant.  That is because agency action must be based on non-arbitrary, ‘relevant factors,’ which here means that the BIA’s approach must be tied, even if loosely, to the purposes of the immigration laws or the appropriate operation of the immigration system.”).

[20] See infra notes 82–84 and accompanying text.

[21] See Davis v. Passman, 442 U.S. 228, 234–35 (1979) (“To withstand scrutiny under the equal protection component of the Fifth Amendment’s Due Process Clause, ‘classifications by gender must serve important governmental objectives and must be substantially related to achievement of those objectives.’” (quoting Craig v. Boren, 429 U.S. 190, 197 (1976))).

[22] See, e.g., Michigan v. E.P.A., 135 S. Ct. 2699, 2707–11 (2015) (holding that the EPA could not lawfully determine that regulation of hazardous air pollutants was “appropriate and necessary” without considering the costs of the regulation where a statute directed the EPA to study the costs associated with the technologies available to control emissions).

[23] Shay, supra note 9.

[24] Cf. Crawford Fitting Co. v. J.T. Gibbons, Inc., 482 U.S. 437, 441–42 (1987) (explaining that where a statute stated that a judge or clerk of any court “may tax as costs” certain enumerated expenses, a Federal Rule of Civil Procedure could not “grant[] courts discretion to tax whatever costs may seem appropriate,” because to do so would “render . . . specific statutory provisions entirely without meaning”).

[25] In other words, Section 385 could recharacterize debt into equity to arrive at an effective tax rate of 45%.

[26] One might argue that Section 385 is limited to corporate debt, but that has not stopped the IRS from proposing regulations that would recharacterize partnership debt.  See 26 CFR § 1.385-3T(f).

[27] For a discussion of the evolution of the concept of statutory purpose in American jurisprudence, see generally John F. Manning, The New Purposivism, 2011 Sup. Ct. Rev. 113; see also John F. Manning, What Divides Textualists from Purposivists, 106 Colum. L. Rev. 70 (2006) (arguing that even though “[m]odern textualists acknowledge that statutory language has meaning only in context, and that judges must consider a range of extratextual evidence to ascertain textual meaning . . . textualism nonetheless remains distinctive because it gives priority to semantic context (evidence about the way a reasonable person uses words) rather than policy context (evidence about the way a reasonable person solves problems)”).

[28] See Landgraf v. USI Film Prods., 511 U.S. 244, 286 (1994) (“Statutes are seldom crafted to pursue a single goal, and compromises necessary to their enactment may require adopting means other than those that would most effectively pursue the main goal.”).

[29] See Ragsdale v. Wolverine World Wide, Inc., 535 U.S. 81, 93–94 (2002) (“Like any key term in an important piece of legislation, the [relevant] figure was the result of compromise between groups with marked but divergent interests in the contested provision… Courts and agencies must respect and give effect to these sorts of compromises.”); W. Va. Univ. Hosps., Inc. v. Casey, 499 U.S. 98–99 (1991) (“The best evidence of [a statute’s] purpose is the statutory text adopted by both Houses of Congress and submitted to the President.”).

[30] See Ragsdale, 535 U.S. at 93–94.

[31] See Motor Vehicle Mfrs. Ass’n of U.S. v. State Farm Mut. Ins. Co., 463 U.S. 29, 42–44 (1983) (“Normally, an agency rule would be arbitrary and capricious if the agency has relied on factors which Congress has not intended it to consider…or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.”).

[32] 26 CFR § 1.956-2T(a)(4).

[33] The 2014 Notice, which originally set forth the deemed U.S. property rule, stated that “Section 956(e) directs the Secretary to prescribe regulations to prevent the avoidance of the provisions of section 956 through reorganizations or otherwise; an inversion is an example of such a transaction.”  Notice 2014-52, 2014-42 I.R.B. 712.

[34] The deemed U.S. property rule is limited to “expatriated foreign subsidiaries.”  A non-inverted foreign-parented group with U.S. corporations and lower-tier CFCs can therefore “circumvent” Section 956 by having its CFCs lend directly to the foreign parent.

[35] See Chevron, 467 U.S. at 842-43 (“When a court reviews an agency’s construction of the statute which it administers, it is confronted with two questions.  First, always, is the question whether Congress has directly spoken to the precise question at issue.  If the intent of Congress is clear, that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress.”).

[36] See § 956(c)(2)(F), (L).

[37] In addition, such transactions could be attached under the circular cash flow doctrine or the anti-conduit theory.  See Rev. Rul. 83-142 (ruling that where a domestic corporation contributes a sum of money to a subsidiary and then immediately receives that same sum as a dividend, the contribution is a “circular flow of cash,” which “is a transitory step that has no federal income tax consequences”); 26 CFR § 1.881-3(a)(3)(ii) (permitting the IRS to disregard an intermediate entity in a financing arrangement for the purpose of determining the non-effectively connected income of a foreign corporation).

[38] See T.D. 9761, 2016-20 I.R.B. 743, 81 FR 20865 (“The Treasury Department and the IRS have concluded that it is not consistent with the purposes of section 7874 to permit a foreign acquiring corporation to reduce the ownership fraction for a domestic entity acquisition by including stock issued in connection with other recent domestic entity acquisitions. Moreover, the Treasury Department and the IRS do not believe that the application of section 7874 in these circumstances should depend on whether there was a demonstrable plan to undertake the subsequent domestic entity acquisition at the time of the prior domestic entity acquisitions.”).

[39] The full text of section 7874(g) reads: “The Secretary shall provide such regulations as are necessary to carry out this section, including regulations providing for such adjustments to the application of this section as are necessary to prevent the avoidance of the purposes of this section, including the avoidance of such purposes through (1) the use of related persons, pass-through or other noncorporate entities, or other intermediaries, or (2) transactions designed to have persons cease to be (or not become) members of expanded affiliated groups or related persons.”

[40] See, e.g., Amanda Athanasiou, Chamber of Commerce, IRS Face Off in Inversion Rule Challenge, 153 Tax Notes 355 (2016) (noting the U.S. Chamber of Commerce’s argument in litigation that the serial acquisition rule is invalid for not “depend[ing] on a scheme intended to circumvent the statute by inflating the foreign corporation’s size”).

[41] For example, a series of acquisitions, none alone which constitutes “substantially all”, but in the aggregate does.

[42] See T.D. 9591, 2012-28 I.R.B. 32, 77 FR 34790 (June 12, 2012) (“[I]f, pursuant to a plan (or series of related transactions), a foreign corporation completes two or more acquisitions described in section 7874(a)(2)(B)(i) involving domestic corporations or partnerships (domestic entities) then, for purposes of section 7874(a)(2)(B)(ii), the acquisitions are treated as a single acquisition and the domestic entities are treated as a single domestic entity.”).

[43] T.D. 9453, 2009-28 I.R.B 114, 74 FR 27922 (June 12, 2009); see also id. (“The preamble to the 2008 final regulations explains that any regulations issued would clarify that references in section 7874(a)(2)(B) to ‘a domestic corporation’ shall, as appropriate, mean ‘one or more domestic corporations’ where the properties of more than one domestic corporation are, directly or indirectly, acquired by a foreign corporation pursuant to the same plan.” (citing 26 CFR §1.368-2(h))).

[44] See Smith v. Commissioner, 78 T.C. 350, 389 (1982) (“The step transaction doctrine generally applies in cases where a taxpayer seeks to get from point A to point D and does so stopping at points B and C in-between.  The whole purpose of the unnecessary stops is to achieve tax consequences differing from those which a direct path from A to D would have produced.  In such a situation, courts are not bound by the twisted path taken by the taxpayer and the intervening stops may be disregarded or rearranged.” (citing Gregory v. Helvering, 293 U.S. 465 (1935))).

[45] In case there was any doubt, the legislative history makes clear that “as may be appropriate” means, or at least should be read as meaning, as may be appropriate “to carry out the purposes of this provision.”  See H.R. Rep. No. 108-755, at 572 (2004) (Conf. Rep.) (“The Treasury Secretary is given authority to issue regulations appropriate to carry out the purposes of this provision and to prevent its abuse.”).

[46] The de-CFC rule was originally set forth in Notice 2014-52 (Oct. 14, 2014).

[47] See H.R. Rep. No. 103-213, at 655 (1993) (“It is intended that the provision apply not solely to back-to-back loan transactions, but also to other financing transactions.  For example, it would be within the proper scope of the provision for the Secretary to issue regulations dealing with multiple-party transactions involving debt guarantees or equity.”).

[48] The IRS has previously exercised its authority under Section 7701(l) to adopt (1) regulations under section 881, see T.D. 8611, 1995-2 C.B. 286, Conduit Arrangement Regulations, 60 Fed. Reg. 40997 (Aug. 11, 1995), and (2) regulations recharacterizing “fast pay stock” arrangements under Treasury Regulations Section 1.7701(l)-3.  The fast pay stock regulations similarly create extra steps and therefore also introduce material questions of authority.  See T.D. 8853, 2000-1 C.B. 377, 65 Fed. Reg. 1310-01 (Jan. 10, 2000); see also Peter M. Daub, The Conduit Regulations Revisited, 147 Tax Notes 409 (2015) at 409.

[49] I.R.C. § 385(a).

[50] I.R.C. § 385(b).

[51] See I.R.C. § 1504(a)(1) (defining “affiliated group” to mean “1 or more chains of includible corporations connected through stock ownership with a common parent corporation which is an includible corporation,” but only if the common parent owns 80% of the total voting power and value of at least one of the other includible corporations and every includible corporation within the chain is owned 80% by vote and value by another includible corporation); Treas. Reg. § 1.385-1(c)(4) (defining “expanded group” using concepts similar to those used to define the term “affiliated group,” such that, in general,  an “expanded group” contains foreign corporations and corporations that are 80% owned by vote or value).

[52] 26 C.F.R. § 1.385-3(b).

[53] 26 C.F.R. § 1.385-3(c)(3)(C).

[54] 26 C.F.R. § 1.385-3(c)(4).

[55] 26 C.F.R. § 1.385-4T(b).

[56] 26 C.F.R. § 1.385-3T(b)(3)(vii)(B).

[57] 26 C.F.R. § 1.385-3(g)(3)(i), (iv).

[58] 26 C.F.R. § 1.385-3(c)(3)(ii)(A).

[59] We focus our analysis on the Preamble to the Proposed 385 Regulations because the Preamble lays out the IRS’s authority argument more clearly.  The Preamble to the Final and Temporary 385 Regulations does respond to comments regarding the IRS’s authority to issue the Section 385 regulations (see “Summary of Comments and Explanation of Revisions”, II.A.), however the response largely repeats the arguments made in the Preamble to the Proposed 385 Regulations.

[60] REG-108060-15 (Apr. 25, 2016), []; 2016-17 I.R.B. 636, 641 (Apr. 25, 2016), []; 81 FR 20916 (Apr. 8, 2016), [].

[61] S. Rep. No. 91-552, at 138 (1969), [].

[62] Among others, the Preamble cites to Kraft Foods Co. v. Commissioner, 232 F.2d 118 (2d Cir. 1956); Arlington Park Jockey Club, Inc. v. Sauber, 262 F.2d 902, 906 (7th Cir. 1959); Uneco, Inc. v. United States, 532 F.2d 1204, 1207 (8th Cir. 1976); Talbot Mills v. Commissioner, 146 F.2d 809 (1st Cir. 1944); and Sayles Finishing Plants, Inc. v. United States, 399 F.2d 214 (Ct. Cl. 1968).

[63] REG-108060-15, 2016-17 I.R.B. 636, 643, 81 FR 20917 (Apr. 8, 2016).

[64] In fact, Section 385 was adopted at the same time Section 279 was added to the Code.  Section 279 limited interest deductions attributable to debt issued in certain corporate acquisitions and the Senate Report, which considered Sections 279 and Section 385 together, made it clear that Congress authorized the IRS to issue general debt/equity guidelines that would cause debt to be treated as equity even if such debt were not subject to the interest deduction limitations Congress set forth in Section 279.

[65] H.R. Conf. Rep. 101-386, at 562 (1989).

[66] H.R. Rep. 102-716 (1992) (emphasis added).

[67] Prop. Treas. Reg. 1.385-2.  A sampling of quotations from the Preamble to the Proposed 385 Regulations (emphasis added in all quotes): “This all-or-nothing approach is particularly problematic in cases where the facts and circumstances surrounding a purported debt instrument provide only slightly more support for characterization of the entire interest as indebtedness than for equity characterization, a situation that is increasingly common in the related-party context.  The Treasury Department and the IRS have determined that the all-or-nothing approach frequently fails to reflect the economic substance of related-party interests that are in form indebtedness and gives rise to inappropriate federal tax consequences.”  REG-108060-15, 2016-17 I.R.B. 636, 639, 81 FR 20914 (Apr. 8, 2016).  “Related-party indebtedness, like indebtedness between unrelated persons, may be respected as indebtedness for federal tax purposes, but only if there is intent to create a true debtor-creditor relationship that results in bona fide indebtedness.” REG-108060-15, 2016-17 I.R.B. 636, 640, 81 FR 20915 (Apr. 8, 2016).  “It is increasingly problematic that there is a lack of guidance prescribing the information and documentation necessary to support the characterization of a purported debt instrument as indebtedness in the related-party context.”  REG-108060-15, 2016-17 I.R.B. 636, 641, 81 FR 20915 (Apr. 8, 2016).  “This requirement also serves to help demonstrate whether there was intent to create a true debtor-creditor relationship that results in bona fide indebtedness.” REG-108060-15, 2016-17 I.R.B. 636, 641, 81 FR 20916 (Apr. 8, 2016).  “These requirements are necessary to the conduct of the multi-factor analysis used in the Mixon and Fin Hay line of cases to determine the nature of an interest as indebtedness for federal tax purposes.” REG-108060-15, 2016-17 I.R.B. 636, 641, 81 FR 20916 (Apr. 8, 2016).

[68] “The Treasury Department and the IRS have determined that the issuance of a related-party debt instrument to acquire stock of a related person is similar in many respects to a distribution of a debt instrument and implicates similar policy considerations.”  REG-108060-15, 2016-17 I.R.B. 636, 643, 81 FR 20917 (Apr. 8, 2016).  “In light of these policy concerns, the proposed regulations treat a debt instrument issued in fact patterns similar to that in Kraft as stock.”  REG-108060-15, 2016-17 I.R.B. 636, 642, 81 FR 20916 (Apr. 8, 2016).

[69] 26 CFR §1.385-3(g)(4); REG-108060-15, 2016-17 I.R.B. 636, 648, 81 FR 20922 (Apr. 8, 2016).  (“Thus, the term debt instrument for purposes of proposed §§1.385-3 and 1.385-4 means an instrument that satisfies the requirements of proposed §§1.385-1 and 1.385-2 and that is indebtedness under general principles of federal income tax law.”).

[70] Lee A. Sheppard, News Analysis: Tailoring the Proposed Debt-Equity Regulations, 152 Tax Notes 311 (2016).

[71] See State Farm, 463 U.S. at 43 (“Normally, an agency rule would be arbitrary and capricious if the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.”).  We think it is fair to say that, all other things being equal, the crazier a rule, the less likely it is to withstand such review.

[72] Going one step further, a regulation that completely reversed all existing case law (e.g., by determining that all factors previously considered to be indicative of debt are in fact indicative of equity) would likewise not stand.

[73] State Farm, 463 U.S. at 43

[74] Utility Air Regulatory Group v. E.P.A., 134 S. Ct. 2427, 2442 (2014) (quoting City of Arlington v. F.C.C., 133 S. Ct. 1863, 1868 (2013)).

[75] Id. (quoting Robinson v. Shell Oil Co., 519 U.S. 337, 341 (1997)).

[76] Kraft Foods Co. v. Commissioner, 232 F.2d 118 (2d Cir. 1956).

[77] Id. at 121–22.

[78] Id. at 124.

[79] See, e.g., id. at 123 (“Undoubtedly, there are elements in the present case which invite close scrutiny by the Commissioner, among them the fact that the arrangement was made between a parent corporation and its wholly-owned subsidiary.”); Nassau Lens Co. v. Commissioner., 308 F.2d 39, 46 (2d Cir. 1962) (“It is all very well to say that it makes little ‘economic’ difference whether the investment [of a parent in a subsidiary] was divided between debt and equity or was entirely allocated to equity.  But by the same reasoning it may make little ‘economic’ difference to [the owner of the parent corporation] whether he has an unincorporated business or a corporation in which he is the sole stockholder.  And if one wanted to carry the argument far enough, presumably the Commissioner would be free to make ad hoc attacks on a whole variety of transactions involving closely held corporations.  But this Code does not so empower the Commissioner, for it recognizes and treats corporations as separate entities and affords significance to the type of investment chosen in them so long as that investment has substantial economic reality in terms of the objective factors which normally surround the type chosen .”); Malone & Hyde, Inc. v. Commissioner, 49 T.C. 575, 578, (1968) (“We start from the premise that arrangements between a parent corporation and its wholly owned subsidiary ‘invite close scrutiny.’  At the same time, we think it unwarranted to apply legalistic and mechanical tests, in the area of parent-subsidiary relationships, without regard to the realities of the business world and the manner in which transactions are handled in the normal and ordinary course of doing business.”); PepsiCo Puerto Rico, Inc. v. Commissioner, T.C. Memo 104 T.C.M. (CCH) 322, 2012 W.L 4207299, at *18 (2012) (“[N]otwithstanding the greater scrutiny afforded to related-party transactions, we believe that disregarding petitioners’ international corporate structure based solely on the entities’ interrelatedness is, without more, unjustified.”).

[80] See, e.g., Kraft, 232 F.2d at 123 (“[T]here are elements in the present case which invite close scrutiny by the Commissioner, among them the fact that the arrangement was made between a parent corporation and its wholly-owned subsidiary.”); cf.  Litton Business Systems, Inc. v. Commissioner, 61 T.C. 367, 377-78 (1973) (“It is quite clear that a valid debt may exist between parties even where no formal debt instrument exists.  This is particularly true in the case of related parties since [f]ormal debt paraphernalia of this type in a closeknit family or corporate cousins are not as necessary to insure repayment as may be the case between unrelated entities.”).

[81] They ultimately rejected this approach because “Treasury believed it would have been unsound policy in effect to deny corporations access to shareholder capital in the form of indebtedness when loans on the same terms could have been obtained from independent lenders.”  45 FR 86440 (Dec. 31, 1980).  These regulations were never effective and were withdrawn.  T.D. 7920, 1983-2 C.B. 69.

[82] See, e.g., Kenco Restaurants, Inc. v. Commissioner, 206 F.3d 588, 595 (6th Cir. 2000) (noting that Section 482 only justifies transfer pricing reallocations if “an arrangement between related parties differs from those reached in an uncontrolled, arm’s-length dealing”).  If the IRS were to adopt a rule recharacterizing all related party debt as equity, we believe it would be appropriate to adopt correlative changes to the Code to account for the economics of third party borrowing in the context of  a consolidated group.  As Debbie Paul noted in her article on July 25, 2016, it is often more efficient to have a single group member be a third party borrower, the creditors of which rely on the equity of the borrower’s affiliates to support the debt.  Related party debt (when issued in appropriate amounts) is therefore a mechanism to allocate interest deductions with the group members that are effectively supporting the third party borrowing.  If related party debt were ignored for tax purposes and no additional changes were made to the Code, the result would be a non-economic allocation of the interest expense 100% to the third party borrower, even if the third party creditors are relying in part on the credit support of lower-tier subsidiaries.  Deborah L. Paul, How to Kraft (or Not Kraft) Debt-Equity Regulations, 152 Tax Notes 525 (2016).

[83] As noted (and rejected) above, one might argue that because the IRS could have taken an even broader approach to recharacterizing debt as equity in the related-party context, the IRS cannot be restricted from adopting a more limited approach.  In other words, the IRS could have declared all related-party debt to be equity and then provided specific exceptions precisely to arrive at the per se rule.  But that argument cannot prevail for the reasons discussed in Part I.  If the broader approach is authorized because the IRS considered factors germane to the question, in order for the “exclusions” to be authorized, the “exclusions” must have the same logical connection to the substantive debt/equity inquiry.

[84] Kraft Foods Co. v. Commissioner, 232 F.2d 118, 126 (2d Cir. 1956).  While the court in Talbot Mills did note in its analysis that “[t]he issuance of the notes brought no new capital into the corporation”, the court, in finding the instrument to be equity, also relied on the fact that the notes (1) were subordinated, (2) paid a variable rate of interest based on profits of the corporation (subject to a cap and a floor), and (3) paid interest annually, but such payment could be deferred by the directors of the corporation.  The court in Sayles Finishing, the other case the IRS cited, states: “Conversion of equity interests into evidence of indebtedness does not necessarily defeat tax treatment of the resulting relationship between a stockholder or stockholders and the corporation as a bona fide indebtedness. Here again, all of the facts and circumstances of the particular case must be considered. Even the fact that no new money was invested in the business does not require a holding that issuance of corporate debentures was a capital transaction, but such debentures are to be allowed tax treatment as indebtedness when the other factors in the case establish a true indebtedness.  [citations omitted]  However, in an appropriate case, lack of new money can be a significant factor in holding a purported indebtedness to be a capital transaction, particularly when the facts otherwise show that the purported indebtedness was merely a continuation of the equity interests allegedly converted.”  Sayles Finishing Plants, Inc. v. United States, 399 F.2d 214, 220 (Ct. Cl. 1968).

[85] We do not believe that any regulation that fails to create tax symmetry between two economically similar or even identical transactions is per se invalid.  Nor should a regulation have to pick up every potential abuse to be upheld.  However, where a specific fact pattern is being offered as evidence of something (in this case, either “equity-like” features or potential for abuse – it is not entirely clear), it is necessary to test whether such a fact pattern is actually relevant to the substantive question, and these examples demonstrate that the no new capital factor is not relevant.  Similarly, the IRS can issue regulations in piecemeal – addressing one factual situation in a set of regulations, with more robust and broad reaching regulations to follow.  That, however, cannot bless every set of “along the way” regulations.  The IRS cannot issue regulations targeting corporations whose name begins with A, for example, and justify the regulations by declaring that the regulations dealing with B-Z are to follow.

[86] REG-108060-15, 2016-17 I.R.B. 636, 642, 81 FR 20917 (Apr. 8, 2016).

[87] REG-108060-15, 2016-17 I.R.B. 636, 641, 81 FR 20916 (Apr. 8, 2016).  While this rule was not retained in the Final and Temporary 385 Regulations because those regulations are limited to debt issued by U.S. corporations, the IRS did not pare back the regulations in response to questions of authority.  Instead, the Final and Temporary 385 Regulations “reserve on their application with respect to debt issued by foreign issuers due to the potential complexity and collateral consequences of applying the regulations in this context where the U.S. tax implications are less direct and of a different nature.” T.D. 9790, 2016-45 I.R.B. 540, 568, 81 FR 72882 (Oct. 21, 2016).

[88] S. Rep. No. 91-552 (1969).

[89] It is too cute to declare this hypothetical regulation unauthorized because the IRS is constrained by Section 385(b)’s reference to factors taken into account in a “particular factual situation” and this example is not a particular situation, but all situations. That linguistic hurdle is easily cleared; we can limit the “all stock is debt” rule to U.S. corporations and non-U.S. corporations with foreign source income.  The limitations of Section 385 must be more fundamental than whether a clever lawyer can wordsmith a regulation so clearly at odds with Congressional intent.

[90]  REG-108060-15, 2016-17 I.R.B. 636, 653, 81 FR 20926-20927 (Apr. 8, 2016).  (referencing H.R. Conf. Rep. No. 2543, 83rd Cong. 2d. Sess. 59 (1954)); see also T.D. 9790, 2016-45 I.R.B. 540, 612, 81 FR 72922 (Oct. 21, 2016).

[91] Preamble to the Final and Temporary 385 Regulations, T.D. 9790, 2016-45 I.R.B. 540, 613; Treatment of Certain Interests in Corporations as Stock or Indebtedness, 81 Fed. Reg. 72858, 72922, (Oct. 21, 2016).

[92] REG-108060-15, 2016-17 I.R.B. 636, 654; Treatment of Certain Interests in Corporations as Stock or Indebtedness, 81 Fed. Reg. 20912, 20926-20927 (Apr. 8, 2016).  In its request for comments, the IRS seems to believe its power extends to recharacterizing partnership preferred equity.  REG-108060-15, 2016-17 I.R.B. 636, 656; Treatment of Certain Interests in Corporations as Stock or Indebtedness, 81 Fed. Reg. 20912, 20929 (Apr. 8, 2016).

[93] “Thus, consistent with the longstanding practice of the Treasury Department and the IRS to apply aggregate treatment to partnerships and their partners when appropriate, and in accordance with the legislative history of subchapter K, the final and temporary regulations generally treat a controlled partnership as an aggregate of its partners in the manner described in the temporary regulations.” T.D. 9790, 2016-45 I.R.B. 540, 613; Treatment of Certain Interests in Corporations as Stock or Indebtedness, 81 Fed. Reg. 72858, 72922 (Oct. 21, 2016).

[94] T.D. 9790, 2016-45 I.R.B. 540, 613; Treatment of Certain Interests in Corporations as Stock or Indebtedness, 81 Fed. Reg. 72858, 72922 (Oct. 21, 2016).

[95]  Id.

[96] For purposes of applying the insolvency exception to recognition of cancellation of indebtedness income to the debt of a partnership, insolvency is tested at the partner level. I.R.C. § 108(d)(6) (2015).

[97] The IRS acknowledges the aggregate theory is being adopted to “prevent avoidance of these rules through the use of partnerships.” REG-108060-15, 2016-17 I.R.B. 636, 653; Treatment of Certain Interests in Corporations as Stock or Indebtedness, 81 Fed. Reg. 20912, 20926-20927 (Apr. 8, 2016).

[98] Applying the aggregate theory would cause a partnership’s debt-equity ratio to include debt owed by its partners, even if not secured by the partnership’s assets.

[99] REG-108060-15, 2016-17 I.R.B. 636, 650; Treatment of Certain Interests in Corporations as Stock or Indebtedness, 81 Fed. Reg. 20912, 20923 (Apr. 8, 2016).

[100] An interesting point to note is that the IRS’s regulatory authority under section 7701(l) can only be exercised “to prevent avoidance of any tax imposed by this title.”  The anti-conduit financing regulation in Treasury Regulations Section 1.881-3 is consistent this limitation. That regulation allows the director of field operations to disregard an intermediate entity only if the financing arrangement significantly reduces the tax that otherwise would have been imposed.  26 C.F.R. 1.881-3(b)(2)(i) (2012).  Consider whether the conduit rule can be used to recharacterize partnership equity as stock if such recharacterization does not reduce the expanded group’s U.S. tax liability.