16 November 2016 Toronto Centre Canada Revenue Agency & Tax Professionals - International Tax Issues

This is an edited transcript of the CRA presentation at the Toronto Centre Tax Professionals seminar held on 16 November 2016. The presenters were:

Paul Stesco, Manager, International Advisory Services Section, International Tax Division, International Large Business and Investigations Branch, CRA

Paul Mulvihill, Partner, Ernst & Young LLP


CRA (Paul Stesco): Our presentation this morning is based on a series of topics that was selected by the committee. These are things that we felt that people wisehd to hear about.

Paul Mulvihill: We all know that you are here to listen to CRA (Paul Stesco), but I will try to supplement. We are certainly not going to talk about all OECD issues and all CRA policies – we will zero in on certain transfer pricing actions relating to transfer pricing Base-Erosion and Profit-Sharing actions – and then a few of the transfer pricing policies and procedures.

Paul, I think you are going to provide some background on the roles of, and interactions between, Head Office and the field.

CRA (Paul Stesco): The first thing that they wanted me to address is BEPS, but I’d like to narrow it down to two specific topics, because these are the topics that I am most familiar with, and they are some of the more important ones. BEPS itself was a G20 OECD project that started in 2013, and the final reports came out October 2015. In total, there were 15 action items, but I’m going to concentrate on:

  • Actions 8-10, which deal with transfer pricing; as well as
  • Action 13, which has been narrowed down to country-by-country reporting, but will address what else is in there.

Actions 8-10: Transfer Pricing

CRA (Paul Stesco): With respect to Actions 8, 9 and 10, which deal with transfer pricing, I think it is important to understand that there are different tacks with respect to this, and people view it differently. I think it’s important to understand the way CRA views it, based on conversations I’ve had since the Actions were published.

I think a lot of the confusion comes from the nomenclature. Terms are being introduced but, when you get down to it, I don’t think it really changes the way CRA has viewed transfer pricing. “Delineating the transactions” is a significant change in the wording and so forth, but I don’t think it alters what CRA has tried to achieve – looking at functions, assets and risk, and doing an appropriate functional analysis.

I think this concept of “delineating” is just clarifying, and making the guidance more robust as to what you should be looking at and how you should be approaching these things. But we don’t consider it to be a significant change in the way we’re going to approach transfer pricing audits, because these are the important things that you have to do in a transfer pricing audit, which is to understand what’s going on. I think there are other elements of the reporting that address that aspect as well.

It’s important, because I think auditors are going to be using these terms, and there are going to be disputes, where people may say “you can’t use BEPS for the years that you’re auditing because BEPS wasn’t in place then.” BEPS is a series of actions that were taken. These specific actions, 8 to 10, address transfer pricing, and they are just a clarification of what we feel CRA has been doing in the past.

Paul Mulvihill: Just to add to that, I think that is CRA’s view, and it has been clear from the budget announcement and other discussions with the CRA, that this is just a clarification of previously established concepts and policies, and just adding further rigour.

I think it is fair to say that there are many other jurisdictions out there for which this is fairly new; Canada isn’t changing its s. 247 legislation, but other jurisdictions may be. It really depends on your jurisdiction and how this may affect reporting in other jurisdictions and the actions by those other tax administrations.

In other tax administrations I think it’s fair to say “this is a new concept – why are we applying it retroactively to prior years when your BEPS legislation is just coming out now?” It’s fair to say that some jurisdictions are wary of using BEPS terminology in their assessments. I know the IRS are trying to refrain from using BEPS terminology, although they’ve had a concept of economic ownership for quite some time – trying to assert taxing positions or rights on the value-creation within a supply-chain.

Entitlement to profits from intangibles

Paul Mulvihill: This is an EY slide (and there are a few EY slides so don’t attribute this to CRA) [Tax Interpretations summary of slide-content]:

1980s 1990s 2014...
Owner Legal Owner Legal owner
Economic owner Controller/functional contributor
Developer Developer Developer/functional contributor
Routine developer
User User Value contributor
Pure user

This is just a look at the changing perceptions to entitlement of profits from intangibles.

Back in the eighties, the concepts were less developed, but tax administrations around the world were focused on trying to determine how to tax certain transactions between multinationals.

  • The owners of intellectual property were the legal owners – they basically got residual returns.
  • There were developers – they were technically no-risk/low-risk (if they were low risk they would need to get a "bump") and they would typically get a cost-plus type of return.
  • And then the users – the would pay a licence-fee for the IP.

In the nineties it migrated. These aren’t hard numbers, but certainly in ’94, the US transfer pricing regulations came out; in ’95, the OECD guidelines came out.

  • The concept of economic ownership became a lot more important – the difference between legal and economic ownership.
    • The economic ownership really revolved around the activities that an economic owner may be performing – so they’re not the legal owner of an IP asset, which can take many forms, but they were doing activities to enhance, or perhaps funding, the IP.
    • Economic ownership became relevant, and attributing returns to economic ownership became relevant, in the nineties.
  • The developer and user didn’t really change during that period (the developer would get a cost-plus return, and the users would pay a licence fee).

Into the 2000s. This is where I have “2014 going forward,” but that isn’t a hard-and-fast date, and I would agree with Paul that CRA has had these concepts, and has been focusing on these BEPS-type issues for some time, so it’s often been my advice that BEPS is a clarification in Canada. I think it remains to be seen how the CRA will transition into BEPS language and concepts.

  • Once again, the legal owner in this context gets next to nothing – a routine risk-free rate of return. That’s an economist’s view. I’m sure there are lawyers who would argue “legal ownership means something – there’s an entitlement there,” and I think it comes down to the terms and conditions surrounding that legal ownership.
  • Economic ownership is broken out into:
    • Controller/functional contributions – they may get a non-routine return for the economic ownership.
    • The investor – they should get a risk-adjusted return, and that’s a pretty wide scope. What would a risk-adjusted return look like for an investor? That could range from a very low-risk financing arrangement to venture capital, which is high risk and high returns if successful.
  • Then we get into the developer model, and breaking that down to:
    • Whether the developer/functional contributor adds value, and maybe that’s just a bigger "plus" on cost-plus.
    • The routine developer would get routine returns.
  • A similar split for a user, and this is a concept that CRA has held near-and-dear to their hearts for quite some time.
    • A marketer, or someone who exploits an IP bundle, could enhance value or contribute value to a product. For example, if your marketing spend on a brand in Canada is above what would typically be contributed by a licensee, it’s fair to say that there’s maybe some expectation of a lower royalty-rate.

Those concepts have been around for a while and they are getting more entrenched, not just in Canada but, more importantly, in juriisdictions that have not focused on these issues in the past.

CRA (Paul Stesco): A comment on that. It’s interesting because you started by saying the legislation hasn not changed. We actually looked at the BEPS project, and specifically Actions 8-10, and we tried to determine whether we think there needs to be a change in the legislation. Basically, we’ve come to the conclusion that we don’t, so you’re not seeing significant changes with respect to the transfer pricing.

However, the guidelines are evolving. We went through a significant change in 2010, and at that time we asked whether the guidelines would have an effect on our legislation. We don’t think it does, because the legislation is based on the arm’s length principle, and all the guidelines do is inform how to apply the arm’s length principle. It is not changing the principle, but rather building on what we’re seeing in practice in industry. That’s why we look at these things, and you have to put it in that context.

Paul Mulvihill: I’ve always liked the idea that the specific wording of the Canadian legislation lines up with the wording of the Treaty – it’s “what would arm’s length parties do” in both Article IX and s. 247.

It’s fair to say that the wording of the Canadian transfer pricing legislation is fairly open. “What would arm’s length parties do” means “is it in the realm of possibilities of what arm’s length parties would do?” I think those things are being discussed in court. It’s a rather subjective point, and I think that’s probably going to be more problematic as you add context to BEPS – adding more wording and defining “value creation” and assigning returns.

We in Canada may eventually move to more prescriptive legislation, like the US, that is a lot more definitive in describing what the arm’s length principles are, rather than just ask what arm’s length parties would do.

Action 8 Guidance/DEMPE

Paul Mulvihill: DEMPE refers to the “Development, Enhancement, Maintenance, Protection, and Exploitation” of intangible assets, as outlined in the OECD. This acronym focuses on the question of what are the functions or elements that may create value in an IP bundle.

Historically, we’ve looked at functions, assets, and risk as the three pillars to define price or return to a particular party. Now we’re also looking at responsibility and accountability for DEMPE functions or other functions.

I think that, when you get into transfer pricing documentation, it needs to be more robust on a going-forward basis – not because of changes in section 247, but because you are going to be providing Country-by-Country Reporting around the world, and there will be more scrutiny. I think you have to better defend or better address your structures and how your value is created within a supply-chain for a product or service.

Just looking at the first element, development of intangible assets (i.e. R&D), there could be funding-control. Within the development of the asset, you should also consider whether there’s funding-control in the development of a particular IP bundle, such as venture capital funding that’s linked to development milestone payments and determinations, which could determine whether the project is a go or no-go. Because a venture capitalist could be involved in funding the intangible bundle, they have control – so there’s a factor in that development, and maybe in the enhancing component from a funding perspective that needs to be addressed. It’s being considered at the OECD. I know Canada is considering risk-free rates of returns. I think a bigger question will be risk-adjusted rates of return, and that has a fairly wide scope.

CRA (Paul Stesco): I think that this is clarifying what you should be looking at. I think that’s what people are considering around the world – trying to get a better appreciation of what should be looked at.

The more advanced taxing authorities have always been looking at these specific items. As for the developing countries, I think that’s what they wanted – they needed more comprehensive guidance (so that DEMPE was talking more to them).

Paul Mulvihill: Just to reinforce – mere legal ownership doesn’t give you automatic returns. There’s a requirement to directly perform control and identity functions.

I think it’s good to label it, but I don’t think DEMPE is a new concept, just a new description of it. I think it’s helpful not just for multinationals, but for tax administrations, to get a better handle on enhancing consistency jurisdiction-to-jurisdiction – admittedly a tall order.

BEPS IP Ownership Continuum

Paul Mulvihill: This slide [not shown] shows the continuum of IPO ownership from a cash box (for a risk-free rate of return) to full ownership entailing full legal ownership, funding obligations and indemnity obligations giving rise to total returns. The functions in the middle of the continuum are clear funding, and legal ownership plus funding, and those are the investor models. They have risk-adjusted rates of return. Risk-adjusted rates of return are an interesting issue.

Something that hasn’t come up in Actions 8-10 is their use of “returns.” They don’t use “expected returns.” That’s when probability comes into play, and factoring in the probability of success.

I think those are key concepts that will probably be fleshed out as we move forward with Actions 8-10 and further define the investor model. Looking at the probability factors for those different funding models will be interesting, and I think there’s wide scope, between the extremes, from risk-free rates of return for a cash box to full ownership, for this investor model.

CRA (Paul Stesco): One of the biggest challenges is defining the IP. You’ve talked about these returns and so forth, but one of the significant points of conflict that I’ve seen in files is actually defining what is the IP, and what is the value-driver, because that’s where there’s sometimes a disconnect.

I think that one of the purposes of the expanded reporting is to identify that, because now it’s letting the taxpayer set out more comprehensively what they identify as the value-drivers in the company, and the intangible assets that create value.

Summing Up re Actions 8-10

CRA (Paul Stesco): To sum up, this is the way the CRA views the revised guidance to Items 8-10 as clarification, and it’s not likely to have a significant effect on our legislation.

The only two things that we feel are changes in direction are:

  • the risk-free rates of return that Paul alluded to – it’s still something the OECD is working on. Is it a same return overall or is it industry related (questions of that nature), and I think there’s a lot of work to be done; and
  • low-value-added services is a concept that’s basically trying to create an easier environment to deal with whether you can get concurrence between the parties on what constitutes this – it will significantly change the nature of the work that you have to perform.

Paul Mulvihill: There may be some debate on risk-free rates of return (typically treasury-bill rates of return). The investor or risk-adjusting funding model would be even more difficult because of its wider scope.

Action 13: Reporting/Documentation

CRA (Paul Stesco): Action 13 deals with documentation. Specifically, there were three things were identified in that Action plan: the master file, the local file, and – everybody’s favourite – the country-by-country reporting.

It’s interesting when you look at this documentation package and you try to relate it to what we just talked about – this whole concept of letting the taxpayer tell their story. I think Canada’s legislation is basically geared that way. If you look at the nature of what we have required in a contemporaneous documentation package, there are specific required elements, but how they are presented is up to the taxpayer.

Again, from our standpoint, we are content at this point with what we have requested in our legislation, and we’ll see where we go with respect to the master file. We’ve obviously signed on with country-by-country reporting, and we’re full steam ahead.

Paul Mulvihill: The purpose is increased transparency. Tax administrations want to be able to see the full value-chain. I think CRA has always had access to that full value-chain, through asking for what’s happening in other jurisdictions further down the supply-chain, and I think that they’ve always had the ability to obtain that information if it pertained to how they should be filing their returns in Canada and attributing income to Canada.

The CbCR spreadsheet will provide an overview of profits, revenue, employees, assets, and taxes paid.

The master and local file will probably get into what is required in Canada. In most jurisdictions, there will be a master file and local file for multinationals. One thing we are discussing with clients is whether the master/local file is a templated kind of analysis that applies to all other distributors around the world, if you are looking at a company that has multiple distributors. The question is, would that suffice for Canadian legislation? We know that s. 247 hasn’t changed. I would say that Canada has a significantly high threshold – we’ll talk about TPRC [Transfer Pricing Review Committee] and documentation penalties a little later – but I would say that Canada requires further analysis. It’s extremely important to describe the parties to the transaction and give enough description of function, assets and risk, and go so far as to get into accountability and responsibility.

It’s certainly going to be challenging for some of our clients who have very large networks to try to bring this down to a jurisdictional entity-by-entity basis.

The three files

CRA (Paul Stesco): The Master File is not currently required in Canada.

We consider the current documentation received through our contemporaneous documentation requirements to be consistent with the information that is to be in the Local File.

As to the new Country-by-Country Reporting, we’ll be following the guidance set out in Action 13. There’s legislation in the House to that effect.

Country-by-Country Reporting

CRA (Paul Stesco): That legislation is consistent with what was in Action 13. The draft legislation was suggested by the OECD. Again, the first year for reporting is 2016. The reporting deadlines are: due within one year etc. as shown on the slide.

The expectation, where Canada is the ultimate parent, and therefore required to prepare the Country-by-Country Reporting, is somewhere between 100 to 120 parents (so that is what we expect to see). Those documents are obviously going to be shared with the other people who have signed on to the mutual protocol. There is an automatic exchange.

There are other countries – the US in particular – that have not signed on to that. CRA and Finance are trying to get an understanding of how we are going to address that from a policy standpoint, and what we expect to see in the first years and so forth. We will be coming up with a policy to deal with circumstances where the other jurisdiction (the US) does not have these rules - but we want to see that information, because that can be critical.

What is this information for? It’s just another piece of risk-assessment. CRA has to go into the field and explain this to our auditors, to ensure that they use it in the appropriate light. We think, initially, that a lot of this information, because of the way it’s exchanged through competent authority, will be centered in Head Office. A lot of what I’ll call “tier-1 risk-assessment” is currently done by Head Office, and this is a tool that will be incorporated into that risk-assessment stage.

Our position on the CbCR reporting is that it’s like any other piece of taxpayer information. We consider that it is not to be made public. There are jurisdictions suggesting that they would make this public – we instead consider that it’s covered under s. 241. Moreover, because it’s being exchanged through a competent authority-process, it’s subject to the same restrictions as any document exchanged that way.

I think there are a lot of people concerned about that aspect of the information – if we are exchanging with a jurisdiction that is considering using this information in another light, how are we going to deal with that? That’s our position on it, and we’ll see what are the other jurisdictions' final decision on this issue; a lot of these policies are in the discussion stage right now. We hope they’ll have something out sooner rather than later, because that’s important for industry.

Paul Mulvihill: Regarding risk-assessing, you could see tax administrations around the world doing headcount-to-profit ratios. You can imagine that they will be looking at whether there is a high volume of intercompany transactions and low profits – these could be triggers.

It’s going to be pretty easy for a tax administration to quickly put together a risk-assessment tool to carve out their “quick hits.”

CRA (Paul Stesco): I think it will identify areas of concern. It’s like anything else when you look at something on the surface. That’s why you have to perform the audit function, because a lot of people will jump to conclusions and say “this just makes no sense at all based on the numbers.” You need to understand the numbers.

Paul Mulvihill: And there are a lot of industries out there that are low-headcount, high-volume. You don’t need a lot of people to generate a lot of returns. And that’s an interesting concept, because headcount is one column in the CbCR reporting that may cause a lot of problems.

I think you’re right – the OECD and many others have said this is a risk-assessment tool – it is not a tool that will identify adjustments. They are not going to use it to say “okay, a headcount of two; $300 million; we have an easy adjustment here.” They may get screened for audit, and then the details will play out and determine whether that’s a reasonable outcome or not.

CRA (Paul Stesco): We have the same problem now when we are auditing as well. There are certain structures, and so forth – we’ll call them “functionally limited entities” – that have significant profits, and we can’t just go out and say they’re not entitled to that profit. We have to go through the process of gathering information and making sure that that position is consistent with the facts and circumstances.

Paul Mulvihill: A number of jurisdictions have signed on to the MCAA. The last time I looked, it was 44, but it’s growing.

CRA (Paul Stesco): Somebody said it was 49.

Paul Mulvihill: And it will continue to grow. Our numbers for Canadian-parented Multinational Entities was 175 (admittedly that could be E&Y looking to expand its potential client-base!). Our data-sources are one thing, but it’s interesting that one source ticked out 2100 MNEs parented in the US, but the US Treasury stated that it’s only 1600.

The better number is probably Paul’s.

CRA (Paul Stesco): Let’s hope so!

Just touching on what I was discussing before – we are preparing education internally so that our auditors respect the nature of this information and use it appropriately.

Where are MNEs now?

Paul Mulvihill: I think one of the first things to do for multinationals is possibly notification. There may not be notification requirements in Canada, but other jurisdictions do want you to notify, by the end of the fiscal year, who your ultimate parent is, and where you will be reporting. I’m not sure if these jurisdictions have a templated form for that, or if it’s just a letter or notification to the competent authority, but this is critical.

What we’ve done is set up templates for clients where we’ve determined what jurisdictions they’re operating in, and then we’ve laid out what is required in each of those jurisdictions. I think all the large accounting firms have these assessment-tools to determine what the reporting requirements are, what deadlines you have to meet in each jurisdiction if you’re a Canadian-based multinational, or a subsidiary of a foreign-based multinational.

I think that right now it would be advisable to sit down and at least get a better understanding of your readiness, feasibility, risk assessment, and the rest of it. Then you get into implementation, notwithstanding that these will not be exchanged until mid-2018. As Paul mentioned, in 2016, Canada will be in a CbCR year, so try to understand what that report will look like. I think it will require a lot of integration between the Tax and Finance people for some of the larger clients out there that have integrated systems.

The endgame is a sustainable report that you’ll be able to produce.

CRA (Paul Stesco): The only thing that I’ll say is: it’s one thing to have an understanding in your head about these requirements, and it’s another to sit down and actually fill out these forms. As people are starting to do that, we are preparing a forum where questions can be asked. There should be information coming on that soon.

Paul Mulvihill: If you’re approaching the reporting threshold but think you may not meet that requirement in 2016, but a spike in sales or revenue could trigger that requirement next year, it’s probably a good time to sit down and assess your international structures – what’s sustainable, what’s not, and a way forward – not just from a reporting perspective, but an overall BEPS view.

International tax audits

CRA (Paul Stesco): I don’t know if it’s evident out in the field, but there was a restructuring within the CRA. Traditionally, we would have the teams go out and conduct the audit. The change, we think, is for the better for the large case-file population, because it really centralizes who you are dealing with, who basically is the large case-file manager. Traditionally, there were specialty areas where the auditors would be reporting to a team-leader in that specialty area.

Those auditors have been reassigned to the large case-file manager, so now there’s a large case-file team, which would obviously include the domestic auditors, as previously, but now also includes the international and avoidance auditors, who will be reporting directly to the case-file manager.

That shapes the audit-team that you’ll be dealing with. How you’re going to interact with them will be different than it traditionally was – International and Avoidance would carry out their audit in the context of their group, and there wasn’t as much communication with the large case-file manager on the international and avoidance issues. That’s what has changed.

Turning to the basic file population, normally you’ll see a similar type of thing; the only difference is that there are designated specialty teams that are more consistent than what you’ve seen in the past, and they will, from an international or avoidance standpoint, not be embedded in those (SME) teams.

With respect to the International Tax Directorate I come from, our structure has not changed – basically, we’re an advisory service, and we’ll be performing our services in a similar context as before. I as a manger have analysts and economists, and the difference is there will be much more significant communication with the large case-file manager, who is ultimately responsible for the file itself.

Role of the economist

CRA (Paul Stesco): The economists are all located in Ottawa – none are in the field, and so forth.

The economists bring a different perspective to the audit itself. The majority of them have no audit experience per se (although they learn it as they go along) but they look at things from a business standpoint. They’re not tax-specialists (although over time they devlop an appreciation for it).

They create the economic reports, where they identify the way to perceive transactions, and what you would expect to see, and taking that and applying it to the legislation and trying to get an appreciation of how things are working and asking “is there an adjustment there?” These reports are integral to the proposal letter because they identify the CRA’s position with respect to the business itself.

They bring a very different perspective to what I’ll call “traditional auditing,” as well as the traditional way of looking at taxation – it changes the dynamic of what, traditionally, you look at in an audit. In an audit, typically you’re going in and trying to verify that things actually happened, and there’s an assumption of the way it should happen. Here, they instead are asking “is this what should be happening,” and that’s a different slant.

I think it’s changed, dramatically, the way that audits are being performed. The nature of the information that the economists are looking for is quite different than what we’ve seen in the past.

Paul Mulvihill: The comment I’d make about the role of the economist is that, as you mentioned, their views are very much entrenched in an economic perspective, and it is just a perspective. I think that’s why you reinforced the concept of a team, that there is a large case manager involved?

CRA (Paul Stesco): Yes.

Paul Mulvihill: There’s a large case-file manager, there’s an auditor involved, and people bring different concepts to the table. An economic report may be the issued, should it be the defining point in an adjustment.

What we talked about earlier in BEPS is that, when it comes to IP, legal ownership is seen as a diminishing component when you talk about BEPS, but I think that there are probably a lot of views here in the room, from a legal perspective, that legal ownership is important, especially when it comes to rights associated with an IP bundle, for the ability to structure transactions and arrange their affairs in a tax-efficient way.

What I think the CRA would say is “that’s all well and fine, but it’s got to follow the value-creation and DEMPE and fit in that model.”

CRA (Paul Stesco): And that’s where all the significant disputes come about – given the fact that you own something, what do you do in the context of your ownership? That’s always going to be an issue. I think that the guidance has clarified that.

I go back to the 2010 guidance and the whole discussion about risk. I think that a lot of these concepts were fleshed out in there, and now we’re just moving it to a different chapter, for lack of a better word, but the concepts haven’t changed. I think the concepts are a lot more consistent with a lot of these economic views – the way the business is run, and the appropriate facts and circumstances.

Paul Mulvihill: I think we could talk all day about this – especially how you assign risk, based on the contracts between entities.

CRA (Paul Stesco): I know what you’re saying, and I think there’s going to be a lot of discussion surrounding that.

CRA Polices

CRA (Paul Stesco): The other thing they wanted to talk about was policies.

As a result of BEPS, we are doing a complete review of all our current policies to determine whether we have to change them. With respect to risk-free rate of return and low-value-added services, we’ll see how those things flesh out - but there will probably be a requirement to change some of our policies.

Generally speaking, this review is similar to the same review we did in 2010 with the revised guidelines that came out then. We put out a TPM which basically recommended we not do a complete overhaul of all our policies at that time. And as I say, we’re still in that stage, and I think that a lot of policies that we have will probably remain consistent, other than the things that I just mentioned.

This is a list of the most recent policies. I was talking about the low value added services and I think that the [TPM-15] intra-group services policy is one that we actually need to go back to and get a real appreciation of now.

TPM-17 The Impact of Government Assistance on Transfer Pricing

CRA (Paul Stesco): The one policy that was asked to be addressed is the impact of government assistance - and, basically, that’s our most recent policy.

The paper itself talks about a cost-plus methodology. It’s becoming a significant issue because of the nature of government assistance in different regions throughout the country.

In circumstances where the entity in Canada is considered to be offering more of a service-type arrangement to a related party (and the cost-plus model is fairly consistent with what we see in a lot of circumstances), where there’s agreement upon the model itself, and the nature of the appropriate methodology to test the price (which is usually a cost-plus methodology), we are saying that, absent proof that government assistance is being shared within the set of comparable transactions (the transactions which are usually comparable), there would be no deduction from or reduction in the cost-base by any government assistance.

I’ve had discussions with people who are taken aback by this. It’s an interesting argument. Again, you look at it from an economic standpoint, and you look at it from a number of different standpoints with respect to the policy in and of itself (i.e., the government assistance policy itself). You’re trying to get an appreciation of "Why is it there, and what would you expect to see?" And I think the jury’s still out.

It’s not an easy thing to deal with and I think there’s more to come on that, but I think that this is the position that we’ve taken now.

Paul Mulvihill: I'm going to throw my two cents in on TPM-17. I have strong views about it, because it applies to issues we’ve dealt with in the past.

I’m not a fan of the way it’s worded. I believe that we needed guidance on this particular issue, but I think it swings strongly in favour of not allowing a sharing of government assistance. It’s that last line, "unless there is reliable evidence that arm's length parties would have done so." The way the TPM is worded I think gives guidance not only to multinationals or to taxpayers, but also give guidance to auditors, and I think the auditors are going to say "yeah, I’m going to err on the side of caution, and I’m going to assess on this basis," so you’re automatically leading to a dispute, I think.

I think it’s a very high threshold. And when you think about it, we are talking about the pricing of goods or services - let’s say it’s services - that are being provided to a foreign-based multinational. They chose Canada because of government assistance. To allow for netting is just recognizing the fact that the foreign-based multinational parent has some view to the government assistance that is being provided in Canada, and would negotiate on that basis. So if they were an actual arm’s-length party, and they took it upon themselves to negotiate, whether they are making running shoes in India or doing pharma R and D in Montreal, they would take it upon themselves, if they are negotiating with an arm’s length party to provide a service, to understand what the cost base of that entity is.

A great example is Walmart. They understand their suppliers, they know their books and records, they know their financials, and when they negotiate with their suppliers, they have a full appreciation of what price they could probably pay for a product that’s being provided by a third party. They don’t want to pay such a low price that they're going to run their supplier out of business, but they don’t want to pay too high a price.

And I think most multinationals have that sense. They are going to locate in a particular jurisdiction due to the incentives there. Making that decision to locate there, I think warrants pricing the good or services on that basis. So I think in cost-base methodology, there’s two variables, the cost base and the cost-plus on that cost-base. I think the way the CRA is viewing this right now is that they are leaning towards one side of that spectrum and that extreme view, and not allowing for sharing. And I know there has been sharing, I think primarily in dispute resolution to resolve a particular issue, but I think it’s going to be a contentious issue call for sure.

CRA (Paul Stesco): It's a contentious issue. Just to comment on it, the decision to locate somewhere because there is a cost-savings makes perfect sense. The question is "what entity should benefit from it," because the multinational is getting the benefit from it from an economic standpoint, in totality. Now, how you start to divvy up where that benefit should be recognized, that’s where I think the big argument is.

Paul Mulvihill: That’s a fair point. The other extreme is giving all the benefit to that foreign multinational. I don’t think that’s reasonable either. So, it’s a fair point, Paul. I just think that sharing is probably the right answer. I don’t have a magic bullet to resolve that right now. Usually, it’s netting or not netting the government incentives and adjusting the rate of return on the "plus" of cost-plus usually does not give you enough juice to get over this hurdle. So, it is a difficult issue.

Business restructurings

CRA (Paul Stesco): I think that what constitutes a business restructuring has changed. And I think that you’re seeing where there’s a change in the way businesses carry on their specific functions is now being looked at a lot more carefully, because that’s basically what a business restructuring is. And I think, historically, CRA did not look at the restructurings through the same lenses that we are now, because changing contracts is a business restructuring, and requires an appreciation of what motivated the contractual change and things of that nature.

So I think that we’re going to be looking at things through the context of business restructuring and, again, consistent with what Chapter 9 of OECD Guidelines talks about. I think that a lot of transactions historically were accepted as "given." However, when you change your contract, was that really a business restructuring? I think that is being looked at through that lens, and I think that there is a lot more audit-work that needs to be done on these, because in some instances, we may have accepted what was going on prior to the change, and we may not have performed a "fulsome audit."

I think there’s a frustration level where taxpayers are saying "Well, that transaction is done - why are you asking questions about it?" However, to understand a business restructuring, you have to have a complete understanding of what went on before, and what went on after. And where the before hasn’t been audited, I think that there’s a taxpayer concern that we're asking questions on what we have looked at before. However, maybe we didn’t look at it before, because the risk analysis was that the transactions seemed appropriate in the circumstances and we may not have done so. These are issues that we are starting to see some pushback on, but you have to be able to appreciate that.


CRA (Paul Stesco): The Transfer Pricing Review Committee is a committee in Ottawa (of which I am a member) and we look at two things:

  • we look at the imposition of penalties under s. 247(3); as well as
  • recharacterizations - that's 247(2)(b) and (d).

Regarding thresholds, we're currently looking at those amounts, and whether there will be a change. As time goes on, we’re starting to see smaller files come in that I don’t think this is what it was geared for. We have to take a better look at that.

Regarding the penalties themselves, what we're trying to identify here is specifically the second part, where the slides talk about 247(3) penalty referrals, is that these are cumulative cases that have come to the committee. So up to 2012, you can see that the incidence of a penalty-recommendation was about 51%, and then subsequently, it’s dropped to 44%.

So there has been a significant decrease in the number of recommended penalties. And I think that’s just a function of the fact that that people are taking the time to conform to the appropriate documentation requirements, so we are seeing things that are a lot more robust and a lot more appropriate in the circumstances. And I think that a lot of the issues initially were that people just were deemed not to have met the requirements, either by not filing anything or the nature of the material wasn’t as good.

Paul Mulvihill: So the "deemed" part probably went down.

CRA (Paul Stesco): Yes. It’s gone down significantly.

Paul Mulvihill: I think it will be interesting to see what will happen in the future. It’s going down, but Canada still has a very high threshold for documentation, right?

CRA (Paul Stesco): The only other thing from that slide is the fact that recharacterizations aren’t done on a regular basis. They are the outliers.