Raymond Stahl of Ernst & Young discusses concerns about the final regulations for claiming foreign tax credits, particularly how they affect creditability, withholding, and treaty benefits.
This transcript has been edited for length and clarity.
David D. Stewart: Welcome to the podcast. I’m David Stewart, editor in chief of Tax Notes Today International. This week: credit where it’s due.
Last December, the Treasury Department and the IRS released the highly anticipated final regulations for claiming foreign tax credits. These FTCs allow a dollar-for-dollar reduction in U.S. tax liability for foreign-paid income taxes. And although the final FTC regulations were released almost six months ago, the rules have sparked quite a discussion within the tax community on how they’ll affect companies seeking to claim the credits.
Tax Notes contributing editor Carrie Brandon Elliot will talk more about that.
Carrie, welcome back to the podcast.
Carrie Brandon Elliot: Thanks, David. It’s good to be here.
David D. Stewart: Could you tell us about these regs and what they do?
Carrie Brandon Elliot: The regs were published in T.D. 9959. As you said, it was released in December of 2021 and published in the final register in early January of this year. The regs are effective as of March 7.
The regs address several international tax sections of the code, but the FTC regs that we’ll be discussing are mainly found in Section 901 and 903. Section 901 allows a credit for income war profits and excess profits taxes paid to foreign countries, whereas Section 903 allows a credit for income taxes, war profits, and excess profit taxes “in lieu of” the Section 901 taxes.
There were also some rules under Section 861-20 for the allocation and apportionment of foreign taxes to Section 904 limitation categories.
David D. Stewart: Now, as I mentioned, these regs have been a hot topic in the tax community, something that both you and our news team have written about extensively. What are some of the concerns you’re hearing about these regs?
Carrie Brandon Elliot: The main concern about these Section 901 and 903 regs is that they curtail creditability of foreign taxes that were previously more easily creditable. That applies to net income taxes where new requirements, such as a net gain requirement and a cost recovery requirement, can curtail a taxpayer’s ability to take FTCs for net income taxes paid to a foreign country.
There’s also concerns regarding withholding taxes, where certain nexus jurisdiction and source rules also can work to curtail a taxpayer’s ability to take a credit for a withholding tax. Another issue is how treaty benefits are affected by these new regs. Also, there are concerns about the allocation enforcement rules in 861-20 that can really adversely impact section 904 FTC limitation calculation.
David D. Stewart: Now, I understand you spoke with someone about these issues. Who did you talk to and what did you discuss?
Carrie Brandon Elliot: Well, I spoke with Ray Stahl, who’s a principal in the National Tax Department at Ernst & Young. He advises clients regarding all aspects of U.S. international taxation, including cross-border acquisitions, dispositions and restructurings, FTCs, inbound taxation, foreign currency transactions, and other matters.
Prior to joining EY in 2020, Ray was a special counsel in the Office of the Associate Chief Counsel International at the IRS. In that role, he was responsible for assisting in the development and implementation of international tax guidance which would’ve included some of these rules.
Ray and I basically spoke about how the new FTC regs affect creditability of net income taxes, withholding taxes, especially royalty taxes. We talked about how these regulations may or may not affect treaty benefits. we also were able to move on to a discussion about section 861-20.
David D. Stewart: All right, let’s go to that interview.
Carrie Brandon Elliot: Welcome, Ray, to the podcast. Today, we are going to talk about foreign taxes and we’re going to focus on creditability, especially the new regs that came out in late last year.
Let’s just jump in. I understand that there are some creditability concerns in the tax community with these new FTC regulations. They exist within the net income tax world as well as the withholding tax world.
Let’s start with the net income tax world and talk about how these regs affect creditability for net income taxes, specifically maybe mentioning the new cost recovery rules. Can you say a few words on that?
Raymond Stahl: Absolutely. Well, first of all, thanks for having me, Carrie. I really appreciate it.
Like you said, I think the regs have really been a little bit of a shock to the system. The regulations were released December 28, and there are a lot of rules in there. But like you said, the biggest part of the package, or the part that’s gathered the most attention, has been the new definition of a creditable foreign income tax.
The rules were billed in part as a response to novel taxes that were being proposed by other countries that effectively expanded their tax base at the expense of the U.S. government, which would give a credit for those extraterritorial taxes. Things like digital services taxes, equalization levies, and the like.
The proposed regulations had the so-called jurisdictional nexus requirement that a tax had to satisfy in order to qualify as a creditable income tax. That was renamed or rebranded the attribution requirement and folded into the net gain requirement under section 901.
But while the government was in there, they sort of undertook what I think of as a lot of deferred maintenance under the section 901 regulations and provided a lot of updates to the section 901 regs. Those updates have led to a lot of unpleasant surprises for a lot of clients who have found that the newly reformulated section 901 net gain requirement, in many cases, leads you to either be concerned that you might not get a credit for what people thought of as a traditional creditable net income tax, or they conclude that they in fact cannot get a credit at all — there’s no path.
What did they do? The cost recovery requirement, like you mentioned, is really one of the two central problems that the government has. But before you even get to the cost recovery requirement, it’s worth kind of stepping back. I always think of it in terms of, first there’s a paradigm shift with the 901 rules and where it shifted from a normal circumstances standard, which was more of a facts-and-circumstances standard in which people could ask whether under normal circumstances does a foreign income tax meet the standard that you’re trying to satisfy?
Empirically, over 40 years of experience with the old regs, people got comfortable that most of our major trading partners’ net income taxes generally satisfied the net gain requirement in the normal circumstances in which the tax applied and it became a non-issue for most income taxes. That’s all out the window now.
What’s been the normal circumstances approach to a section 901 analysis is replaced with an objective test where you’re supposed to be able to mechanically analyze, get a translation of the foreign tax law, and analyze whether, by its terms, the foreign tax law complies with a series of specific, objective requirements.
You can understand why, from the government’s perspective, when it’s difficult to obtain data, it’s difficult to actually look behind how tax laws are applying in their normal circumstances, why that would offer some appeal that they would want a more objective test.
The problem is the specific elements of the objective test that are under the net gain requirement are very tight. Obviously, the world is complicated and there are many countries with many different tax systems. As you start getting into the different rules in each of these different systems and you’re trying to reconcile this extremely mechanical, somewhat unforgiving standard with the various different tax systems in the world, you find that it’s difficult to conclude with a huge degree of comfort that any of these tax systems actually satisfy these mechanical tests. The cost recovery requirement is, I think, the best example.
Big picture, what is the cost recovery requirement? What is it about? It says that a foreign tax law has to allow for the recovery of significant costs and expenses. It has a per se rule that tells you that certain costs and expenses are significant in all circumstances. That per se list is extremely long. It’s all capital expenditures, interest, rents, royalties, wages, research and experimentation, payments for services.
At the end of the day, you realize we’ve quickly transitioned from a rule where we’re asking, “On balance, does a foreign tax system generally allow expense recovery against gross income?” Now we’re in a rule where your starting point is that you ask, “Is every dollar of most of the expenses that we think of as traditional business expenses recoverable?”
Fortunately, it doesn’t stop there. You keep going. There are some exceptions. The key exception being: foreign tax law generally is treated as allowing recovery of significant cost and expenses. As long as the denial of the recovery is based on principles that are similar to a denial that underlies a federal income tax denial.
The challenge then is every single time that you find a limitation on the recovery of costs or expenses in a foreign tax law. There are obviously hundreds and hundreds in our major trading partners, and they’re all different. They obviously don’t start with the IRC when they pass their tax law. You have to ask, “Is the principle that underlies this disallowance similar to the principle that underlies some corresponding disallowance in the United States?” The challenge then is, “How broadly do I interpret that language about principles?”
The regulations have somewhat confusing language around how you should think about those principles. Because on the one hand they say look for disallowances that limit base erosion or profit shifting, but then they go on to give an example of a type of disallowance that works. It’s essentially a rule that sounds a lot like section 163(j), but set at 10 percent of a measure for taxable income rather than 30 percent.
I always think of that as the two poles. You can say, “Hey, on the one hand, I can read the one sentence that tells me I’m just looking for a rule that’s motivated by BEPS.” Then you get to the next sentence and you say, “Well, wait a minute, maybe the rule has to actually operate like a U.S. disallowance provision or deferral provision, or maybe it needs to bear some similarity. There needs to be some rule of reason that tells you that the disallowance has to be reasonable in some context.”
To give you an example, one of the rules that people get nervous about in the base erosion context is in Hong Kong, generally speaking, interest payments to foreign-related parties are not deductible except in certain circumstances. Hong Kong also doesn’t withhold on those interest payments.
So, on the one hand you step back and you say, “This seems like a perfectly reasonable rule. It’s motivated by the concern that interest expense, among related parties, could be used to strip the Hong Kong tax base, particularly in the absence of any type of withholding tax.”
On the other hand, you realize, “Hey, you don’t need any indicia of base erosion to be present. They could be paid to a high-tax-related party. It could be a payment. It could be $1 of interest expense on a company that doesn’t have any sort of thin cap issues.”
You begin to wonder, “Well, wait a minute, we don’t have any rules in the United States that operate on a hair trigger like that to prevent base erosion. Does this look enough like 163(j), or the base erosion and antiabuse tax, or other anti-base erosion measures in the United States in order for the Hong Kong interest expense disallowance to be treated as satisfying the cost recovery requirement?”
That’s just one example. I think a lot of countries have their own unique ways of preventing base erosion.
What we’ve been doing is essentially trying to go through every country’s tax rules, understand the way they operate, and then try to reconcile them or compare them to provisions in the U.S. code, which is quite an exercise.
Carrie Brandon Elliot: Wow. It sounds like it’s not just a compliance requirement, it’s an uncertainty problem, especially when it comes to cost recovery.
Aside from cost recovery, which sounds like a primary obstacle, you had mentioned there were some other troubling aspects of the approach to net income taxes in creditability rules of the regs. What else is there besides cost recovery?
Raymond Stahl: There are a lot of other things, I would say. I think cost recovery is the first huge problem. The second huge problem that is in the regs is the royalty sourcing rule.
Carrie Brandon Elliot: We’re going to pivot to withholding tax then as opposed to net income tax?
Raymond Stahl: Yes. There are other problems and issues in net income tax. It’s a good point.
I should mention there are also four elements to the net gain requirement. Realization requirement, the gross receipts requirement, the cost recovery requirement, and an attribution requirement.
Those first two, realization and gross receipts, don’t present as many problems, but they do present some problems, and it’s worth keeping in mind that the government may look for ways to potentially soften the blow that these regs have landed. If all of the attention goes to cost recovery, it’s worth bearing in mind that there are some countries where the country’s tax systems look like they potentially present problems under the realization in gross receipts requirement.
Solving that kind of issue where you have very tight, all-or-nothing type rules that apply mechanically and then knowing that there are just many different timing and realization regimes out there, I think that’s going to be a real challenge for the government. If the focus ends up being on cost recovery and sourcing going forward, will we still have major problems under those other elements?
Carrie Brandon Elliot: Let’s move to sourcing then. I understand that there’s been a lot of concern about withholding taxes and sourcing, specifically with regard to royalties.
Raymond Stahl: The fourth element of the net gain requirement is the new attribution requirement. Generally speaking, what that requires is that if a tax is imposed on a nonresident, it has to meet one of three tests in order to satisfy the attribution requirement. Typically, when you are looking at a withholding tax — withholding tax on dividends, interest, rents, royalties imposed on nonresident — typically there are a lot of elements to the analysis. But what you’re usually focused on is whether or not the gross income that is in that withholding tax base is sourced is subject to tax by a reason of it being sourced under the local country’s rules in a manner that is reasonably similar to the sourcing rules that are under the IRC.
The easiest example I always think of is being services. The U.S. rules generally determine the source of services based on where the service is performed. If I’m sitting in the United States, and I’m providing a service, and I’m paid for it, that’s going to be U.S. sourcing.
If I’m providing that service to someone who’s in another country, and that country withholds by reason of the fact that the customer was in the other country, they say that service is sourced locally, then that tax is not going to be creditable because they’ve determined the source is using the location of the customer rather than where the services are performed.
My suspicion is that’s very much intentional. The point of this rule is that the government feels that when you don’t follow traditional sourcing norms, then that’s a way of expanding the local country’s tax base and potentially getting the U.S. tax credit to offset that. The goal was in part to limit countries’ ability to grab tax that really should be collected by the United States.
The biggest problem that’s presented has been related to royalty withholding taxes. The rules have a special rule for royalty withholding taxes. They say that the gross income from a royalty must be sourced by reference to the place of use of the intellectual property.
I don’t think a lot of people have ever really had to focus on why a country was withholding on a royalty stream before, but what we’ve found as we’ve begun to survey the royalty withholding taxes that exist out there is that royalty withholding rules are a lot more complicated than simply just withholding on the basis of place of use.
In many instances, and I would say potentially almost as common as, if not more common than a place of use rule, is a residence-based rule. If a royalty is paid by a resident of Country X to a resident of Country Y, then Country X will just withhold. I believe that’s the rule in China, France, and many other major trading partners. That’s an oversimplification as the way the rule works. It can get more complicated, but often it starts with a resident-based sourcing rule. Then there can be additions and exceptions.
The regulations are very clear. They have an example telling you that a resident-based royalty withholding tax doesn’t satisfy the attribution requirement. You step back and you say, “OK, this is a rule that’s premised on the idea that you don’t get a credit for a tax that doesn’t satisfy traditional international taxing norms and the idea being that the U.S. code represents those norms.” But in fact, what we found is that maybe the U.S. code isn’t really consistent with international taxing norms when it comes to royalty sourcing.
You have a situation where a lot of companies that suffer huge royalty withholding tax liability are now looking at this liability. They’ve been claiming credits for just very large numbers for many years in what they think of as a vanilla creditable tax and they’re potentially going to lose those credits. To make matters worse, that can be the case even if in fact the IP is only used in the jurisdiction that is withholding.
The rules are very clear about the determination as to whether or not a tax is creditable is based on the terms of the law and not the facts on the ground.
There’s an example in the regulations that shows that when a resident of one country licensed IP to a resident of another country, and the licensee country collected withholding tax, it didn’t matter in that example that the IP was actually only used in the country they withheld. They deny a credit because the withholding was based on the residence of the payer, the residence of the payee.
That is, I think, a huge pressure point. There’s probably enormous sums of withholding taxes that are caught by this rule. A lot of taxpayers are very frustrated with this rule and are likely talking to Treasury, the IRS, and Congress about that.
Carrie Brandon Elliot: The irony of it is if you wanted to plan around that sort of thing, it would be a lot easier if you had intercompany licensing arrangements than if you had actual arm’s-length licenses with unrelated parties. It’s almost as if taxpayers are in a bind on this if they do a lot of non-intercompany licensing and royalty paying. How are you advising your clients on this?
Raymond Stahl: Yeah, that’s a great point. I think a lot of advisers have started to explore ways to plan around this. Obviously, first of all, it’s not ideal if you are restructuring, potentially incurring even more foreign tax than you otherwise would have to structure into a net income tax, rather than pay a royalty-withholding tax so as to ensure that you get a credit.
But you’re right, when it’s an internal transaction, there are potentially more options the taxpayers have going forward in terms of eliminating or mitigating the extent to which they’re losing credits for loan taxes. Whereas third party licenses, it’s more difficult to restructure those in a way that you can avoid a withholding tax or that you can claim a credit.
Carrie Brandon Elliot: Since we’re talking about source rules and withholding taxes, we’ve got to talk about treaties. How do these rules affect the treaty situation?
Raymond Stahl: Yeah. That’s, I think, one of the most interesting aspects of the mess that we’re in.
The regs say, “If you’re entitled to a credit under the relief from double tax article of a treaty, then that tax is creditable, even though it wouldn’t have satisfied the requirements of the reg.”
In the preamble, Treasury and the IRS sort of acknowledge this. They say, “Well, by regulation we wouldn’t be able to take away what we’ve given, what we’ve granted in a treaty.”
For example, let’s say you had a royalty withholding tax on a royalty that was paid to a U.S. domestic corporation that qualified for treaty benefits, and under the terms of the treaty was entitled to a credit for that withholding tax. Even if that withholding tax would fail the sourcing requirement under the attribution rules, then you’d still be entitled to a credit under the treaty.
But then the preamble of the regulations goes on and says, “But wait a minute, you only get a credit to the extent that you’re eligible to claim a benefit under the treaty.”
Take that same example where there’s a royalty received, say by a controlled foreign corporation of a domestic corporation. The royalty is subject to a withholding tax and that withholding tax isn’t creditable under the regs. Because the CFC is not a domestic corporation that’s eligible for treaty benefits, it can’t claim a credit under the treaty.
But the regs are conspicuously silent and the preamble is also conspicuously silent on whether or not a deemed paid credit would be available under the treaty.
I think the easiest way to think through it is to look at it in terms of the German treaty. The German trade tax has limits on interest expense and other things like rents and royalties, I believe, with embedded financing elements that are clearly not motivated by anti-base erosion measures.
For a lot of advisers, it’s either a very close call or it’s difficult to get comfortable with the idea that you can get a credit under the regs for the German trade tax. You say, “But wait a minute, the German treaty expressly allows a credit for the German trade tax.” I think, “OK, great, I’ll just claim it under the treaty if I have a domestic corporation with a German CFC that pays trade tax.”
What Treasury and the IRS have said publicly on panels is, “Not so fast. The double tax relief article of the treaty allows a credit essentially for deemed-paid taxes on dividends.” In fact, the technical explanation even refers to section 902. The argument goes, “Well, that’s moot. Section 902 has been repealed in connection with the Tax Cuts and Jobs Act and there’s essentially no longer a treaty credit for deemed-paid taxes.”
Some advisers differ. Some advisers are saying, “Well, wait a minute, not so fast. I need to read the treaty in light of the changing statutory rules that are in place in each country over time. Also, while giving effect to the intent of the treaty partners when they signed the treaty and global intangible low-taxed income has replaced our deferral system, section 960(d), which allows a credit for GILTI taxes as a successor to section 902. Maybe I need to read the treaty really flexibly and interpret the treaty so as to treat the GILTI inclusion potentially as equivalent to a dividend and allow a 960 credit under the treaty.”
This is an area where there’s been some pretty clear and direct messaging from the government on panels that they don’t agree with that position. I think a lot of advisors are nonetheless going to take that position and taxpayers are going to take that position. Potentially as a litigation matter going down the road, because it’s not something that they can really deal with, as they’ve sort of admitted in the preambles. If it turns out that there is a credit, I don’t think you can issue a regulation to deny the credit, so it’ll be interesting to see how that plays out.
Carrie Brandon Elliot: Right. Assuming that at foreign tax survives all of the hurdles of creditability, it looks to me like the regs add yet another hurdle. And that is section 861-20, where we allocate an apportion foreign tax expense to limitation categories.
Can you talk about how that layers over on the rest of the situation?
Raymond Stahl: Yeah, that’s been a little bit of a sleeper issue. We found clients are all on different ends of the spectrum in terms of where they are in terms of 861-20 awareness.
There are some clients that have realized, “There are huge issues here and I’ve got tons of lost credits by raising the 861-20 regulations.” There are other clients that are just kind of waking up to the compliance challenges. Once you get through those compliance challenges, the traps that are there in the regulations.
Unlike the 901, 903 regulations, which are perspective, or at least for calendar year taxpayers only went into effect for 2022, for the most part, the 861-20 regulations are retroactive. And for most calendar year taxpayers apply going back to 2020.
There’s tons of latent liabilities out there, but very big picture: The biggest trap that we’re seeing that comes up most often is that, under the disregarded payment rules, the disregarded payment rules are intended to allocate taxes to 904 categories or to income items for 960 purposes when the taxes are imposed on a disregarded payment, which generally makes sense because that had been sort of a missing piece in the allocation picture before the regs were issued.
But the way in particular that the regulations deal with taxes imposed on so-called “disregarded” and so-called “remittances” — which are really disregarded dividend income for the most part, but it can also be other payments that are treated like remittances — is that what the rules do is rather than link the taxes to current year income or previously taxed earnings and profits that the CFC may have, which is more likely to result in the taxpayer being able to claim a credit, the regs use essentially tax basis as a proxy for current year income.
That might not sound that problematic, but you can have circumstances where you have, for example, a CFC owns a disregarded entity (DRE). The DRE has a lot of cash and the cash is treated as a passive category asset. The DRE makes a distribution of that cash subject to withholding tax and the taxes overwhelmingly allocated to the passive category income item at that CFC level. Then the CFC in turn doesn’t have enough passive category income in order for there to be a deemed-paid credit, and the taxes are just lost.
It may have tons of tested income that year, but it doesn’t have a lot of basis because it’s assets are fully amortized or depreciated. We’re seeing that over and over again. It’s really caused massive sums to be lost.
I think that rule is causing a lot of heartburn as people begin to realize that it’s there and apply it.
Carrie Brandon Elliot: Do you get a sense that the U.S. government is aware of and in tune with some of the problems we’ve identified? Do you have any sense of how things are going to go forward?
You briefly mentioned that you knew that taxpayers were contacting government and there was some dialogue between industry and the IRS and Treasury. What is going on out there?
Raymond Stahl: Yeah. I think it’s clear that Treasury and the IRS are aware that they have a problem and they’ve been responsive and want to fix it. For a while, there have been rumors about what form that fix might take: a technical correction, or a proposed regulation, some combination. We’ve recently heard from the government on panels that they’ve identified sympathetic cases and it sounds like they’ve got at least a two-track approach based on those comments.
First on the royalty sourcing, they’ve acknowledged that there are sympathetic cases out there where you may have a license to a person that exploits the IP only within a particular country and yet, because the country imposes a residence test, essentially to determine whether or not to withhold on that royalty, there’s no credit. One of the things that the government referred to on a panel recently was the addition through a proposed regulation of a safe harbor that looks for indicia of use within that country.
I believe that one thing that was mentioned was if the license itself limits the use to a particular country, then the country’s withholding tax will be deemed to be imposed on the basis of use. I think that’s a good start. Like you were saying earlier, it’s a lot easier for taxpayers to renegotiate internal licenses in some cases and provide that the use of the IP is limited to a particular country.
Now, query whether that’s true in all cases, it’s difficult under federal income tax principles to determine where IP is exploited in a lot of fact patterns. As they renegotiate those licenses internally, they’ll have to get comfortable or see if they can get comfortable with the idea that they are using the IP within a particular country.
But that would be one area that would provide some relief for internal licenses. I think it’s going to be more challenging. That safe harbor alone may not provide much relief for third-party licenses, which are often regional or territorial and provide rights to exploit IP over and within several countries.
We haven’t heard whether or not the government would be willing to expand that relief, but it seems worthwhile considering those are true third-party commercial arrangements in many cases.
Carrie Brandon Elliot: That’s interesting. I know that creditability of foreign tax is one of the most important aspects of international tax. If you curtail or remove it, it creates a lot of anxiety.
With that, I think that we’re out of time and it’s time to sign off. I really thank you for appearing today. I think we’ve had a great, useful, and informative discussion.
Raymond Stahl: Thank you for having me, Carrie.