Oral argument — Altera Corp. v. CIR

Judge Graber and I are very pleased to be joined again by Judge O'Malley from the Federal Circuit Court of Appeals and who made a special trip out here for this case and we really appreciate you taking the time to fly in to hear this important matter. Thank you. Nice to see all of you again so you may proceed. May it please the Court, Arthur Catterall of the Justice Department on behalf of the Commissioner and I'd like to reserve five minutes for rebuttal if that's okay. This case is about whether Treasury permissibly concluded that the 2003 cost-sharing amendments to the Section 482 regulations are consistent with the Arms Link Standard and whether it adequately explained the basis for its conclusion during the rulemaking process. Now because the Arms Link Standard is actually articulated in and made applicable in every case by the Section 482 regulations rather than Section 482 itself, we view the permissibility inquiry as comprising two issues which we set forth in our supplemental brief. The first issue is whether Treasury's understanding of the Arms Link Standard as that understanding is reflected in the 2003 cost-sharing amendments comports with the language of the regulation articulating that standard which is Section 1.482-1B1 and the second issue is whether Treasury's understanding of the Arms Link Standard comports with the authority granted to Treasury under Section 482 to allocate income and deductions among commonly controlled entities in order to prevent evasion of taxes or clearly to reflect the income of such entities. Okay let me let me just try to make sure that I'm clear on how you think we structure this this analysis. Putting aside the State Farm issue and where it would show up in the inquiry, isn't the first question whether or not the statute itself is ambiguous? In other words permitting regulation in the space. That's you referring to step one of Chevron. Right. Yes. All right and you agree I mean I went back and listened to the argument from the last time you agree the government agrees that the statute is ambiguous is that right? Right. In fact I think it does not address the precise question. Okay I think your words were there's no doubt that it's ambiguous as the way you phrased it the last time. Okay. All right and and and that's because there's nothing that explicitly says that the commencement with income standard should could be applied in these circumstances is that right? Correct and and I mean yeah there's there's nothing in the statute that addresses cost-sharing arrangements in general and also as you know the the words arm's length do not appear in the statute so we think it's ambiguous on well it doesn't but but you can see that it's implicit in yes okay. I have a slightly different question if I may. Looking directly at the statute the first question seems to me to be whether the arrangement here is a transfer within the meaning of section 482 which is not about the right the other regulations but it's just about the statute and if the cost-sharing arrangements are transfers do they also have to be arm's length or is that a separate permission? No they have they have to be arm's length. Why? Because the regulation expressly says that. But you take a different view of what arm's length means. That's this whole this whole case is about what does the arm's length standard mean and and who gets to say what it means. I understand how with respect to transfers you made the argument or that that Congress passed a statute that says as it relates to transfers and licenses it's possible that the comparables are not perfect. I mean was wasn't that what at least the the initial purpose of the amendment was? I think the purpose of the amendment was to it basically creates a presumption that that transactions involving intangible property between unrelated parties the evidence is... But the language of the statute doesn't say transactions right it says transfers and licenses. Correct. And there are comparables when it comes to transfers and licenses correct? I would I would say that's the whole point behind the commensurate with income requirement is that Congress did not think that the reliable enough. Right but I guess I'm I'm trying to get to the point is that there is a distinction between an imperfect comparable and the total absence of a comparable correct? Sure sure. And so with respect to transfers and licenses there are always comparables they're just what Congress felt is that they were imperfect. Sure sure I would agree with that. Okay now you've taken the position that even when I mean is your position that there are no comparables with respect to cost-sharing arrangements or that even if the comparable is that it wouldn't be taken into account that fact can be ignored? Well what we're saying is that basically the commensurate with income requirement establishes a presumption that unless the transaction the the transaction that you're and unless the transaction the uncontrolled transaction unless that transaction if you're trying to you know infer a result from the if the result you get does not satisfy the commensurate with income requirement then what the alleged comparable is simply deemed not to be comparable. Well what do you what's your response to your friend's argument on the other side that if you take that to its logical extreme you do away with comparability analysis completely because you can always just say well let's just use commensurate with income if we don't get an exact parity or we don't get to where we think the best tax value is. Well I don't think you you do away with comparability altogether it's just that again if you are relying on a comparable transaction in the transfer license context if the result you get by looking at that transaction does not satisfy the commensurate with income requirement then it's simply not a comparable transaction. So you view it as a essentially a definition of what is an arm's-length value in the case in which there is a transfer or license of intangible property it's sort of a subset in your view. Right I mean again the statute is clear that the income attributable to the intangible must be commensurate with the you know with the with the royalty payment. So again if you if you don't satisfy that requirement then it's simply not a comparable. Is there an economic basis for your argument that you can simply define this to be arm's-length or are you just saying that that Treasury has the right to artificially define it to be arm's-length? Well I don't think it's artificial I think again when you look at the legislative history of the of the 1986 Act and Congress said that the the purpose of the commensurate with income requirement was to ensure that that the relative that the that the terms of the transaction reflect the relative economic activity of each party and so that's the economic principle that's the animating economic principle and in practice you know one consequence of the commensurate with income requirement is this notion that in the in the transfer or in the license context you know the idea that you need to go back and make a profit experience and the thing is you know everybody agrees because and and that's in the House report where the you know Congress said we expect this periodic adjustments to be taken and the thing is you know we have cited sources indicating that this periodic adjustment rule does not reflect what parties do in the real world and so you know obviously I mean if you if the commensurate with income requirement has any meaning at all you know it's in other words if under Altair's position you know the arm's-length standard says you have to look at evidence of third-party behavior in all instances well if that were true then evidence of third-party behavior could override this periodic adjustment rule and and nobody says that's the nobody's arguing that so it's the same type of analysis when you when you get to to caution arrangements okay back to the way you wanted to structure the argument you said that other question is whether Treasury's interpretation of the regulations is reasonable the reason I the reason I say that is because again a big part of Altair's argument is that the term you know arm's-length standard you know means X means you have to take into account evidence of third-party behavior in all circumstances and you know the the actual language of the arm's-length standard is in the regulation so it seems to me that you know the first inquiry is whether Treasury is reading its own regulation correctly so we're dropping all the way to our deference is that right well we didn't I mean we didn't ask for our deference I don't think you you need our deference well if we're considering whether Treasury is reading its own regulations in a reasonable fashion isn't that exactly what our is all about well sure and I think you know if I guess ours kind of like Chevron asks you know is this a permissible reading and we certainly think it's permissible but but I wouldn't limit it to a permissible reading we think it's it's the you know it's the best reading of the of the regulation so right so we but the best reading the best reading analysis is applies to the statute right assuming that we have to interpret it in the first instance if we find that there's something inappropriate about the regulation or how it was passed then we decide what the best reading of the statute is in the first instance right but when you're talking about interpretation of the regulation our says is it reasonable interpretation sure sure I do not disagree with you so and so in Xilinx this court said that the regulation the regulatory scheme as whole was was itself ambiguous right right and I know that you have you pointed out to the footnote that they said maybe that the amendments could fix that ambiguity so is that your view that this that by putting these amendments in these coordinating amendments that you've taken away any ambiguity in the regulatory scheme since all the all the pieces and parts have to be considered together yes yes I mean these these amendments make explicit what Treasury was arguing was implicit in the in the prior regulations and they make explicit simply by the ipsy-dixit of saying that this is the arm's-length standard well yeah I mean you know and they and they ground that conclusion on what was in the 1986 conference report which said you know we are aware of these cost-sharing arrangements we don't intend to preclude the continued use of cost-sharing arrangements to the extent that they reflect the principles of this new statutory commensurate with income requirement and then the report went a step further and said you know here's what we mean by that you have to share all costs in proportion to anticipated benefits and that's that's exactly that's exactly what what the regulation says so if I may getting back to the to the to the APA question what's what is amazing to me is that in the 417 pages of comments to the proposed regulations that Altera has provided to the court the 1986 conference report shows up exactly twice and both times it's in a background discussion with no suggestion that Treasury misread it and and one commenter stated in one breath that the proposed regs are not consistent with the arm's-length standard and then it comments the question of whether the proposed regulations are consistent with the congressional mandate under section 482 but if the regulations are consistent with the congressional mandate under section 482 and the mandate incorporates the arm's-length standard then how could the regulations violate that standard or more to the point Congress's understanding of that Thank you, Your Honor. May it please the court, Donald Falk for Altera. Every time the government explains this rule it seems to have a little bit different story and there's a reason for that because no matter what angle they come at it with it just doesn't hold up. Well counsel, clearly the statute requires that if there's a transfer or license of intangible property the income with respect to it must under the wording of the statute be commensurate with the income attributable to the intangible so if this is a transfer or license of income principle why isn't this a transfer? Well I'll answer that question first and then I'll I have another point I think to make that I don't think it even whether or not it's a transfer I don't think that answers the question here it's not a transfer because the cost-sharing agreement allocates in advance the rights of the intellectual property as it's developed and economically it's quite different from a royalty or a license and the But it's about future royalties or licenses for sure because I mean it's otherwise there'd be no nothing to share. But there is no there there is it is something that that affects the economic rights as they come into being and indeed the Treasury recognizes that. It affects them by transferring them or licensing them. Well I don't I don't think that may ultimately be the case in particular arrangements but the the Treasury regulations recognize and in 7 AH1 that that these are reimbursement payments not royalty payments which is significant given the withholding consequences and and and reflects Treasury's recognition also expressed in the white paper the cost-sharing avoids the issues with transfers of intellectual property and the legislative history I think supports the view that the act does not directly apply except in in royalty payment situations. If there is a set is a transfer that is other than a license it may not be in the form of a royalty it could be an outright payment. It could be an outright payment. Or some other form of income. That is that is true but but but what the legislative history says at page 638 of the conference report is cost-sharing arrangements to produce results consistent with the changes made by the act royalty arrangements and and it says they should be proportionate to profit but it doesn't directly apply it doesn't directly apply the standard and in fact I think the key point here is nothing about commensurate with income wipes away the arm's length standard from every stage of the analysis. But vice versa is true also because if an arm's length analysis yields something that is not commensurate with income for a transfer or a license of intangible property then it wipes out the second sentence of section 482. So it's a problem both ways is it not? Your Honor I don't think it's a problem and I think the way the Treasury implemented the the commensurate with income section and in the dash 4f regulations it reflects exactly what what Congress had in mind and the way Treasury understood it as it is said time after time including in this rulemaking that the commensurate with income standard is consistent with the arm's length standard. You begin with an arm's length analysis to figure out what what you know what the what the income is going to be and what costs are allocated and if the cost if the profits are outstripping the the payments and the whatever whatever form the cost-sharing takes whatever form the transfer takes then it can be adjusted by periodic adjustments which is exactly what the legislative history says but the legislative history also says that for cost-sharing that you're looking at actual economic activity which is an analysis of the inputs not the outputs and it's also says in the house report that that that explains how this actually works it talks about all the facts and circumstances on on page 426 of report 99 426 which is the arms length analysis you look at all the facts and circumstances. But isn't the government's position that arm's length and comparability are not one in the same? Well what the government's position is basically that the arm's length standard has nothing to do with arm's length evidence in this narrow area which I think is completely unsupported. No what they're saying is that that we've defined this by regulation to be an we're not using comparability but we've defined it to be arm's length. Why isn't why aren't they allowed to do that? They're not they they're not allowed to do it because this isn't through the looking-glass they don't get to say a word means whatever they say they want it to mean because there is a they they know and and the tax court knew and the Supreme Court knew that the whole power that is given under section 482 is not limitless it is only a power to bring related parties into parity with unrelated parties and since the beginning that has been recognized to reflect an arm's length analysis which is to say not necessarily a perfect transaction but you look at what third parties do you shape whatever the allocation is based on what third parties do and even the legislative history of section 42 says that. In your view what do you do if there are no similar transactions by unrelated parties? There are absolutely no similar transaction in any respect. Right what is your view of how that has to be done? Well you you'd have to have some kind of economic analysis using you know rigorous economic principles. And what they did and let me just follow that up. Sure. If if it and I know you don't agree with this but if it is a transfer or license of intangible property Congress has told us what to do told the Treasury Department what to do hasn't it? It's told the Treasury Department to make periodic adjustments if the income is is is is not in accord with the way the deal was cut if if the too little money is flowing from one side to the other in accord with the benefits. It does not say throw out the arm's length analysis and just sort of start from scratch and have a purely internal analysis and certainly Treasury said nothing of the sort in the rulemaking. What Treasury said was that yes we're invoking commensurate with income. It's consistent with the arm's length standard as we have said for lo these many years. And then when challenged on the point instead of saying oh we don't have to worry about comparables we don't have to worry about evidence. They said no we don't like your evidence so we're gonna make up a hypothetical of our own. But that's that's not discarding evidence of comparable transactions and and I think the important foundational point here is that the tax court found that the evidence of third-party conduct here was probative and included comparable transactions and and the Treasury did not challenge that here. They have not challenged the tax courts finding on its grounds. They have instead said that we have the power to redefine the arm's length standard at will. Which is another thing that that the 15 tax experts in the tax court couldn't find in in any of the notices or preamble that that they could redefine the arm's length standard so that it didn't include arm's length evidence. None of that is there. What if their notices and preambles expressly said that? What if they said we're we're giving you notice we're gonna pass a regulation and we're gonna completely redefine what arm's length means and the way we get there. In other words we're gonna discard comparability analysis. We are going to say that that we believe that commensurate with income is consistent with arm's length standard. If it was very clear on its face from the beginning would they have the authority to do that? Well I think ultimately they do not have the authority to throw parity analysis out the window. The Supreme Court said that that parity is the purpose of section 42 and the addition of commensurate with income there's no indication that it's meant to override that. In fact it's it is meant to enforce parity because of in this in the context of high-profit intangibles which again is the only rationale given. High-profit intangibles where stock-based compensation is significant. The only rationale they give is for this narrow piece of the rule but the but section 42 was not none of it says let's throw out all evidence of what parties do. It says we look at commensurate with income which like everything else is determined using third-party evidence. So I think at a certain point if they said we're throwing out all evidence in all stages of the commensurate with income analysis we're just going to make it up ex cathedra as if we had the power to engage in the sort of formulary apportionment that that Professor Aviona favors. You don't see it here. That's why they're grasping at its straws because they realize that's what they're asking for. They're grasping at 1962 committee reports run enacted legislation has the word formula in it. So there that's where they say they have the power but it's not there. It's not there. I don't think that it would survive. I think they could do a lot of things. I mean I think they do have some flexibility but I don't think they could throw out all evidence of third-party contact when they're all the ultimate goal is to get them to parity get related parties to parity with unrelated parties and and the second sentence solves a very narrow problem. It doesn't throw out the analysis. So I don't think good that would be a violation of Chevron for all the reasons that we said. It's a statute. Are you saying that the regulations are an unreasonable interpretation of the statute or that that they're not entitled to our deference because they're engaging in an unreasonable interpretation of the regulations? Well they're not entitled to our deference because they haven't claimed it and and and the if you're putting regulation up against regulation Ipsodixit doesn't solve the problem in Xilinx. So I don't think that's going to get them our deference or get them anywhere. No we think it's an unreasonable application of the statute for the reasons that we explain at length. The the statute includes a parity analysis and and you know more more to the point or as to the point this whole notion that you can discard all evidence of third parties conduct take the arm's length straight out of the arm's length standard that they recognize as part of the statute that that's nowhere. I have a sort of a separate question that if I may judge I'm sorry. Just you you've been arguing alternatively or two things one has to do with the terms of the regulation and whether it's reasonable but the other has to do with evidence in the record. I thought that the inquiry that we're here to make is the validity of the regulation and not the specific evidence in this case whether it does or does not satisfy the regulation. But in the first instance we're looking at the validity of the regulation and as to that why would the specific evidence in this case make any difference one way or another. I don't think the evidence in this case makes any difference Your Honor. I think the evidence in this case has that this is just a case and you know and in our view they could apply this regulation apply the arm's length standard and they may there may be specific things in this case and indeed there are other issues in this case that affect pieces of the outcome but what they can't do is is ignore the rulemaking record and not meaningfully respond to it which is. What is your response to the government's argument that the rulemaking record may very well show that that these kinds of stock-based compensations is not considered an economic cost but it is it is an accounting cost and so why can't it be considered and maybe they just rejected the the what they thought to be irrelevant evidence. Well the fact that the regulation on its face disavows reliance on financial accounting standards I think ought to take care of that argument as a tax court found and they did not engage with the economic cost argument that was another thing that a tax court noted that there were fairly significant analyses why stock-based compensation is not the kind of cost that is shared by unrelated parties and if 482 gives them the power only to allocate to bring related parties into parity you can't allocate to make people do something unrelated parties can't do. Yes they can go measure it with income that's a very special specialized piece of the analysis. It's more than can counsel it's more than can they must according to the statute. Absolutely your honor we don't dispute that if that applied they they would have to apply it here but that does not allow them to completely throw out arm's-length evidence as to the rest of the analysis their own regulations begin in the the 4F regulations begin with an arm's-length analysis then they figure out whether there needs to be any retroactive adjustment based on profit it doesn't just start it doesn't start in the ether the way they're asking to do here and in the whole point of the of the section 482 adjustment is again it's let the chips fall where they may that's what first security says the the Commissioner comes in and says well I know that arm's-length parties couldn't actually pay this but we're gonna require it to be transferred in the non arm's-length context because we think they can do it and they have different incentives and the Supreme Court says forget about it you can't you can't do that doesn't matter what the result is it matters your power here is to bring parties into parity not to do whatever might produce the most revenue and that that's that's the same thing here no one related parties share stock-based compensation they can reach their analysis and again we don't think it applies to cost sharing we don't think commensurate with income applies to cost sharing but if it did they could they could draw their analysis based on the usual arm's-length factors you take it as far as you can go and then you adjust for the differences in the particular transaction which is what which is what the courts and the IRS have been doing for years I mean that's what all the all their analyses their their alternate methods under commensurate with income look to various types of third-party conduct and and and they figure out what the price ought to have been and and and that's how that's how it works it doesn't start from a vacuum the way they're claiming they can do now and they're claiming you know to be able to get away completely from any kind of empirical analysis and yet they're still claiming that the regulation and that commensurate with income is consistent with the arm's-length standard and as the tax court which knows a thing or two about the arm's length standard having first articulated it in 1933 its predecessor said that's necessarily an empirical determination whether it's consistent with the arm's length standard I may have some leeway as of course they do when it comes to what the numbers are in a particular case but it's an empirical determination well that's what your response to the government's argument is even if it is an empirical determination and even if the tax court does have expertise in that area it doesn't have expertise in Chevron it does not have any particular expertise in Chevron no but it does have it does have expertise in the meaning of the arm's length standard does have expertise in what reasonable experts could discern from the rulemaking record and none of those 15 tax experts could see this argument that they're raising now if they can brush aside that the the arm's length standard the arm's length evidence in the arm's length standard that's the piece that's that's new there now they're coming back to saying they're they're applying the arm's length standard but they're still getting rid of all the evidence and and and no one could see that neither the commenters nor nor the 15 tax court judges could find their new position that evidence is irrelevant or extraneous and and indeed that's not what they said if there were any doubt even if there were some ambiguity even if you know there were some crumbs in the in the notice and I don't think there are that take this this strong a position the preamble should have removed all doubt it talks about the arm's length standard 30 times goes piece by piece of the evidence it's conclusory but it goes piece by piece of the evidence of similar transactions and and then it makes up a transaction of its own saying this is what arm's length parties generally would do the tax court said you can't say what parties generally would do on the all the evidence is to the contrary you have to try something else if you want to get there and they haven't challenged that and that I mean I think is one of the keys here because they are making a taking a position here that is so at odds with what they've said before it's not we're not disputing that one commensurate with income applies they get to look at the income in the particular transaction and adjust the proportions what we are saying is you can't you can't start by defining income in a way that has nothing to do with the arm's length standard in the way related parties define income and put things into income and put things into costs that that unrelated parties would not do they didn't claim that right in the preamble and and they don't have that right I think it's it's also telling that that they at oral argument now they've peeled things back a little bit but they get further and further out in their claims of power from being able to modify the arm's length standard to saying it's in the statute but commensurate with income overrides it the supplemental brief said it never had any content so it doesn't matter what we do and in all this gets back to the most fundamental points here which are which are that you know that it's very important to keep agency rulemaking transparent so regular party regulated parties don't have to guess at the reasoning of agencies particularly when agencies are seeking to expand their power as as the Treasury and IRS are seeking to do here and take the 482 adjustment power beyond the parity purpose that that is recognized to be the purpose of the statute and it's something the IRS can't change the regulation or any other way and that's there's reasons are so that that the regulated parties can comment intelligently can respond can understand without guessing at the agency's reasoning that they can respond intelligently that the agencies can take the input and have their reasoning as well as the mere text of the proposed regulation challenged because the resupporting reasoning is is is critical here because it suggests further powers to do further things as as would any preamble when you are supporting a regulation and also to let the political process work if the agency stakes out an extreme position as it does here and essentially taking the arm's length out of arm's length the arms like standard that's not empty proceduralism it's a critical tool to limiting executive action in a democracy and that's why the tax court made an unusual unanimous 15 to nothing statement to the IRS that you have to follow the APA just like everybody else does and you have to follow it the same way everybody else does and it's especially important for this court to set down the same rules because this court's precedent covers all the agencies not just the Treasury not just the IRS but EPA DHS and LRB obviously you do this this court sees them all these rules apply to everybody's that these principles have to be honored for the IRS just like they have to be honored for everybody else have to say what they're doing and let people respond in a meaningful way and not come back at the end with something and say was hidden under the dust there thank you thank you counsel your rebuttal one thing that struck me when I went back and listened to the first argument is that mr. Falk used twice used the phrase contradiction in terms to describe the notion that some related party transactions must satisfy a purely internal standard in order to achieve an arm's-length result so let me make four points in response to that first the statutory commensurate with income requirement is by its terms a purely internal standard it requires a comparison of the two sides of the related party transaction not a comparison of the related party transaction to transactions between unrelated parties there's simply no getting around that second the commensurate with income requirement is just that a requirement as judge Graber was was noting you know shall be commensurate one one related parties benefit shall be commensurate with the other related parties benefit the third point is that the commensurate with income requirement does not require Treasury to allow taxpayers to override its terms by resort to evidence of third-party behavior and again you can compare that approach with the approach of the proposed overhaul and of section 482 in 1962 when Congress did provide such an escape hatch the fourth point is that we know that two years after it enacted the commensurate with income requirement Congress legislative legislatively confirmed that that requirement is consistent with u.s. tax treaties almost all of which incorporate the arm's-length standard so let me ask you something I forgot to ask you in your first session and that is one of the arguments that you and your amici make is that this is really important because this stock-based compensation has become a real thing in recent times and so there is a tax avoidance that is that is allowed through that mechanism if we don't adjust these costs this way and I guess my problem with that is that you know not that avoiding taxes as long as it's legal is not in and of itself per se bad under the law but but if it's true that this is a whole new thing and that the world is changing how could we say that that was the backdrop against which Congress was was legislating back in 1986 well I don't I don't think Congress was was legislating against the backdrop of stock-based compensation that's I mean that's that's not part of the legislative history you know what they were legislating or the backdrop that they were legislating against was that comparability analysis was not working in the context of transfers and licenses of intangible property so and if I can just get back to the point about the contradiction in terms what Alterra refers to as a contradiction in terms is simply the juxtaposition of the commensurate with income requirement and the arm's-length standard and and clearly Alterra views that as a contradiction in terms because its understanding of the arm's-length standard as always requiring resort to evidence of third-party behavior cannot be reconciled with the commensurate with income requirement so in getting back to the to the APA argument Alterra does not really contend that Treasury failed to was implementing legislative intent as revealed in the 1986 conference report their their real complaint seems to be that Treasury did not adequately explain the rationale behind the legislative intent and in other words they seem to argue that Treasury did not adequately explain how Congress arrived at the conclusion that the purely internal standard of the commensurate with short answer to that argument is that any perceived shortcoming in Treasury's explanation of Congress's rationale for the directive that Treasury was plainly implementing simply cannot form the basis for an APA violation. Thank you counsel. The case is argued to be submitted for decision. I thank all of you for traveling out here for the arguments and we will be in recess for the afternoon.