Individual Seeks Expansion of Regs on Sourcing of Cloud Income

October 6, 2019

Internal Revenue Service
P.O. Box 7604
Ben Franklin Station
Washington, D.C. 20044

Attn: CC:PA:LPD:PR (Notice 2019-30) Room 5203
Mr. LG “Chip” Harter
Mr. Doug Poms
Mr. Willian W. “Wade” Sutton
Mr. Robert Z. Kelley

Re: Notice 2019-30, 2019-2020 Priority Guidance Plan — Sourcing of Cloud Services Income

Dear Sirs:

Preliminary Comments

Source is important to domestic taxpayers because they are concerned with applying the foreign tax credit (FTC) limitation mechanism. Source is also important to foreign taxpayers (including foreign subsidiaries of U.S. taxpayers) because they are concerned with the U.S. taxation of their effectively connected income (ECI). Despite source rules being important to these two classes of taxpayers for different reasons, the source rules themselves are normally applied to each class in the same manner. Thus, if a taxpayer, whether domestic or foreign, earns income from services, then the existing source rules found in Reg §1.861-4 apply.

With domestic taxpayers being concerned with the FTC limitation and the increasing likelihood of foreign countries either unilaterally or through future international cooperation and consensus imposing tax on cloud services where users and customers are located, there is good reason to now consider an approach to sourcing cloud services income. That is necessary to assure such domestic taxpayers an ability to claim appropriate FTCs for foreign taxes that were generally not previously imposed. (Where a treaty applies and allows taxation of cloud income, its provisions may cause the income subjected to foreign tax to be foreign source in applying the FTC limitation. For such treaty-protected taxpayers, new sourcing rules would be of lesser concern.)

At first blush, it would be expected that most U.S. taxpayers would be concerned only with their ability to use foreign tax credits against U.S. tax imposed on their cloud services income. However, it seems likely that the bulk of the cloud service earnings of U.S. multinationals (U.S. MNCs) is earned not through domestic group members, but through their CFCs or other foreign group members. This often results from a commonly-found format of profit-shifting. While I am not aware of (and have not searched for) any specific research on this, a perusal of a few of the Form 10-Ks issued by some of the major U.S. MNCs and the information released in connection with several Senate Permanent Subcommittee on Investigations (PSI) hearings of prior years certainly supports an expectation that the vast majority of cloud service earnings (dollar wise) are within foreign group members of U.S. MNCs. Due to this tax-motivated structuring, the cloud service revenues earned through these CFCs or other foreign group members will typically have, in the past, been subjected to zero or very low levels of foreign tax. With the initiation of taxation by increasing numbers of countries where users or customers are located (whether unilaterally or from some future international consensus), the quantum of foreign tax paid should increase by a material percentage. This creates a critical need for appropriate sourcing rules for income from cloud services that are fair to taxpayers and that do not cause the U.S. Treasury to bear through the FTC mechanism more than an appropriate share of the economic costs of these increased foreign taxes.

MNC Analysis of Domestic and Foreign Cloud Service Structures

Given that MNCs are likely earning the bulk of their cloud services income provided to foreign customers through their CFCs or other foreign group members, their concerns about income source are dual:

(In this paper, I have assumed for simplicity that any taxes imposed by countries where foreign group members are organized or operate, or where they have customers or users, will qualify as credible foreign income taxes even if imposed on a gross receipts basis.)

Any MNC that conducts a cloud service business will require clear sourcing guidance if it is to be able to review its options for the future structuring of its business. Many MNCs will have as alternatives:

• The use of a foreign structure under which a CFC or other foreign group member holds relevant intangible rights for foreign markets and contracts directly with customers and users within those foreign markets (this foreign structure appears to be commonly used by many if not most MNCs), or

• The use of a domestic structure with domestically-held intangible rights and domestic group members contracting with customers and users worldwide.

In the case of a foreign structure as described, there will also normally be a parallel domestic structure that holds rights for the domestic U.S. market and that contracts directly with customers and users within that market. In some cases, MNCs may include within the domestic structure one or more western hemisphere countries.

In analyzing and comparing these alternative structures, a number of factors must be considered by any MNC. These factors might include:

• Varying level of foreign taxes

  A domestic structure should have no tax residency in any foreign country that could cause foreign tax to be imposed based on corporate residency. On the other hand, a domestic structure could include foreign branches (including disregarded entities), thereby causing some taxation on local branch operations. Further, of course, increasing numbers of countries may impose tax based on the location of the user or customer.

  A foreign structure would normally have tax residency in some location, though many MNCs effectively avoid any residency-based taxation on their foreign structures through the use of tax havens and disregarded entity planning.

  A potential difference in the level of foreign taxes between domestic and foreign structures may arise from the relative tax treaty benefits of each structure. For example, a domestic structure will have the benefits of the U.S. network of treaties as well as the strength and energy of the U.S. competent authority in the event of any disputes. On the other hand, a foreign structure's treaty benefits will depend on the residency of the group member that is reporting the cloud services income. Information from the PSI hearings and other sources such as some of the European Commission's State Aid decisions and user agreements available online indicate the use of such countries as Ireland, Luxembourg, Netherlands, Switzerland, etc. Such countries do clearly have treaty networks. It seems, though, unlikely that the competent authorities of these countries would make too much effort or fight too hard for MNCs that are only using their respective countries as a cog within a global profit-shifting structure. (Interestingly, this expectation about how much effort a competent authority might exert is belied by the apparent effort that countries like Ireland, the Netherlands, and Luxembourg have made in connection with State Aid decisions. I see this State Aid dispute as being different, though, from a treaty competent authority proceeding since the former involves a significant effort to assert local taxation control and sovereignty and avoid central EU oversight. Further, since the State Aid decisions are made public in comparison to private and confidential tax disputes concerning a tax treaty or transfer pricing, it's important to these countries to demonstrate to their current and potential foreign investors that they will fight to sustain the tax benefits that they provide through private rulings or other means.)

• Varying level of U.S. taxes

  Income earned through a foreign structure is subject to potential U.S. taxation under one of two regimes. The first is indirectly through the subpart F and GILTI mechanisms. The second is directly through the ECI rules. On the other hand, a domestic structure is taxable under normal rules at the 21% corporate rate. Where the cloud services provided constitute a FDDEI service (proposed Reg §1.250(b)-5), then the effective tax rate should be lower than the normal 21% due to the §250 FDII deduction allowed.

  In any analysis that an MNC makes, it will have to consider for the foreign structure the level of foreign taxes and how much of those foreign taxes will be credible after application of 20% haircut of §960(d) for foreign taxes related to GILTI and the §904 FTC limitation. Due the effect of §904(h) (which is made applicable to GILTI by §951A(f)(1)(A)), the source of income at the CFC level will determine the source of the subpart F or GILTI income inclusion. This creates a critical need for appropriate sourcing rules for income from cloud services.

This new source rule [Reg §1.861-18] would be a significant change for those vendors who have language in agreements with the customer specifying that title passes at a point prior to complete download of the content. Thus, under the proposed regulations, vendors who are located outside the United States but who sell to US customers may find themselves with significant US-source sales income and may, if the proposed regulations enter into effect, have to determine whether they are engaged in the conduct of a trade or business in the United States and therefore owe US income tax on those sales.

Critical Need for Guidance on Source

There is a need to provide all taxpayers with guidance on the sourcing of income from cloud services income. Doing so will not only help taxpayers. It should also help the government's ongoing efforts to deal with some profit-shifting structures.

To date, it appears that the government's approach to attacking profit-shifting structures has mostly been through transfer pricing adjustments to the value of the intangibles that have been transferred into zero- or low-taxed CFCs. There have also been efforts to adjust the computations for sharing development costs among the participants in a cost sharing agreement (CSA). These approaches deal either with a one-time transfer or make a refinement, albeit a very material refinement, of an annual payment under the CSA. Yes, these approaches will result in significant adjustments if sustained. However, they leave in place as legitimate income of the applicable foreign group members very material ongoing cloud service income (as well as some other types of income) that in many cases is measured in the billions, tens of billions, or more on a cumulative basis in the years after the initial transfer of intangible property. These approaches do not deal with the issue that the bulk of the annual cloud service (and other types of) earnings result directly from basic and crucial business activities conducted primarily within the U.S. and not, for the most part, in any location where the foreign group members are tax resident or have personnel, offices, or other physical facilities.

While I am not privy to the tax filings of any MNCs that earn income from cloud services, it seems very likely that in both pre- and post-TCJA years, MNCs have been taking the position in their FTC limitation calculations that their cloud service earnings are fully foreign source in the hands of their foreign group members. This foreign source treatment, which ignores the above-mentioned basic and crucial business activities taking place in the U.S. within U.S. group members, would likely be based on a position that these U.S. group members are acting as independent contractors and not as agents or partners. Under §904(h), this full foreign source treatment in the hands of a CFC affects the sourcing of dividends (pre-TCJA only) and subpart F and GILTI inclusions in the hands of a U.S. shareholder.

While MNCs might have much or most of their cloud service income within their foreign group members, small and mid-sized domestic taxpayers may earn such income within domestic group members. If they are subjected to foreign taxes, as will likely occur more frequently in the future, then income sourcing rules and their effect on the FTC limitation will have an important and significant economic effect on them.

The evolving discussion and application of unilateral taxation of cloud income and the possible international cooperation and consensus to this effect has included both gross receipts-based taxes and tax based on group-wide profitability. It has also focused on the taxing rights of the countries of users and customers. The existing sourcing rules for services were written decades ago, long before the advent of cloud businesses and modern international tax structuring using hybrid entities and other mechanisms to achieve the dual goals of minimized foreign and domestic taxation. These factors make the issuance of cloud income sourcing rules of critical importance.

Expansion of Proposed Reg §1.861-19 and Reg §1.861-4

Coverage of Reg §1.861-19

Recently proposed Reg §1.861-19 defines and provides rules for classifying cloud transactions as either a lease or the provision of services. In separate email communications with Treasury, I have noted that this proposed regulation does not explicitly include as cloud transactions some common cloud-based business models. Several important and material models not explicitly included are:

• Advertising models where end-users obtain “free” services and advertisers pay for access to those end-users,

• Marketplace sites and apps that function as sales agents,

• Gig-economy sites and apps that put service providers and customers together,

• Job recruiting sites and apps that find candidates for employers,

• Travel sites and apps that act like sales agents for hotels, flights, etc., and

• Game sites that allow users access to a range of games for a subscription price.

While I believe it should be obvious that such business models involve cloud transactions that generate income from the provision of services, I understand that there can be alternative opinions on the nature and classification of such models. Accordingly, I believe that it would be very useful to add guidance to proposed Reg §1.861-19 to make clear that all such business models involve cloud transactions that are classified as the provision of services. Perhaps this clarity could be achieved through placing additional examples into proposed Reg §1.861-19(d).

Inclusion of Cloud Services Sourcing Rule within Reg §§1.861-4

Code §§861(a)(3) and 862(a)(3) along with Reg §§1.861-4 and 1.862-1(a)(1)(iii) have long stood for the principle that services are sourced where the services are performed using a facts and circumstances approach. Where services are performed partly within and partly without the United States, Reg §§1.862-1(a)(1) and (a)(7) both provide:

For the treatment of compensation for labor or personal services performed partly within the United States and partly without the United States, see paragraph (b) of § 1.861-4.

Section 863(b) provides the Secretary with the authority to prescribe rules that would apportion taxable income in certain cases. Such rules could use “processes or formulas of general apportionment”. Since the Secretary holds this authority for “gains, profits, and income . . . from services rendered partly within and partly without the United States”, the Treasury and IRS appear to have wide latitude to consider practical approaches to sourcing cloud services income that are generally consistent with the approaches being currently discussed by the OECD Inclusive Framework and other interested parties.

I suggest that Reg §1.861-4 be expanded to include rules and examples for cloud transactions. Such rules and examples would importantly recognize that cloud transaction income that is treated as income from services under proposed Reg §1.861-19 will be sourced based on the facts and circumstances approach that has been included within Reg §1.861-4 and case law for many years. (In this discussion, because of the above-quoted reference in Reg §I.862-1(a), I see Reg §1.861-4 as an appropriate and convenient location for a cloud service sourcing rule. It seems sensible to have all “service” sourcing rules in one place. However, given the above-noted authority granted to the Secretary under §863(b), such a rule could also be appropriately included in regulations under §863.)

Unitary Application of Reg §§1.861-4 to Cloud Service Businesses

New source rules for cloud service income in Reg §§1.861-4 should recognize and value the various factors that contribute to an MNC or other taxpayer's conduct and profitability of its cloud-based business. Given the nature of today's cloud businesses that operate worldwide in a borderless fashion through the internet and typically through centrally controlled and managed multiple group members, there should be a rebuttable presumption that source must be determined on a unitary basis that takes into account the activities of all group members, and where relevant, the activities of cloud-service users. Only where an MNC or other taxpayer can establish that it has two or more cloud businesses that are operated through separate managements, separate operating personnel, and separate physical facilities would the presumption of one business operated on a group-wide basis be rebutted. Where rebutted, source would be determined for each separate business on a unitary basis for each such business. Such a unitary approach is consistent with some of the proposals currently under discussion in the international dialogue being conducted by the above-mentioned Inclusive Framework and other parties.

For each of the one or multiple unitary cloud service businesses, the percentages of U.S. and foreign source income as determined (see below) would be applied to the income recorded within each group member. Where a consolidated U.S. tax return is filed, which should commonly be the case, application at the U.S. group-member level will not normally present any issues for federal taxation purposes. Where foreign group members are involved, the portions of their income that are, respectively, U.S. source and foreign source would affect the U.S. shareholder's FTC limitation calculation under §904(h) to the extent of subpart F and GILTI inclusions. Alternatively, where a foreign group member is found to be engaged in a U.S. trade or business, its U.S. source income will be taxable as ECI. (See submission in response to Notice 2019-30 dated June 2, 2019 regarding the 2019-2020 Priority Guidance Plan from Jeffery Kadet and David Koontz. Appendix A includes suggestions regarding clarification of when a U.S. trade or business will be considered to exist for modern business models, including those that conduct cloud service businesses.)

As an example of the above, assume an MNC conducting a cloud service business through two group members: one established in the U.S. (X) and one established outside the U.S. (Y). Under a cost sharing agreement, each of X and Y owns the business's intangible property necessary to exploit the business within its own territory, which is the U.S. for X and the rest of the world for Y. Management and other personnel responsible for the business's DEMPE functions (development, enhancement, maintenance, protection, and exploitation — see further discussion of DEMPE functions below) are primarily within X in the U.S. Both X and Y have sales, marketing, and customer service personnel within their respective regions as well as relevant tangible assets (e.g. servers, data centers, office equipment, etc.). Under the above described rebuttable presumption, X and Y are treated as a unitary group conducting one cloud service business. From applying on a unitary basis applicable sourcing rules (a suggested approach is described below), it is found that 70% of the business's income is sourced in the U.S. and 30% is sourced outside the U.S. When computing the relevant foreign source income that will be used in X's FTC limitation calculation, 30% of X's own directly earned cloud service income will be foreign source. Further, in applying §904(h) for the source of subpart F and GILTI inclusions, 30% of Y's directly earned cloud service income will be foreign source. If it is found that Y is engaged in a trade or business within the U.S., then 70% of Y's cloud service income will be ECI.

Reg §§1.861-4 Sourcing Rule for Cloud Service Income

The current international dialogue noted above is also considering the reality that in some businesses, especially cloud service businesses, users play a significant role in adding to content, value, and profitability. Some discussion and proposals go further and look to include sales and/or marketing intangibles that may exist even in the absence of important user participation. Given this direction and the likelihood that any consensus reached (as well as any unilateral actions in the absence of consensus) will include factors for user participation, sales, and/or marketing intangibles, Treasury and the IRS should now be thinking about whether any new Reg §1.861-4 sourcing rule for cloud service income should take account of any or all of these factors, which of course go beyond the traditional focus on the location where services are performed.

The current Reg §1.861-4 facts and circumstances approach focuses on the performance of the services and not on “the residence of the payer, the place in which the contract for service was made, or the place or time of payment”. This approach was applied to cloud services in Reg §1.937-3(e), Example 5, where there was a relatively vanilla factual situation described in which all activities and servers were in one location and there was no user content that added to the value of the platform. The example appropriately concluded that that one location was the income source, even though there were customers in other locations around the world.

In today's multinational environment, MNCs conduct unitary cloud service businesses that have customers worldwide and operate through personnel, assets, and offices in many countries. Many small and mid-sized companies and groups have customers and users internationally, even if all of their personnel and assets are located within one country. To determine where such taxpayers earn their cloud services income, it is necessary to take relevant personnel and assets into account (on a unitary basis where multiple group members are involved). Further, as noted above, consideration should be given to also taking customers/users, sales, and/or marketing intangibles into account.

In considering personnel, the following functional categories may be relevant:

As for assets, there will of course be relevant property including computer equipment, office facilities, data centers, servers, etc.

From a customer and user perspective, there are various measurable factors including sales, data content contributed, etc. There are presently efforts being made by many countries in the OECD's Inclusive Framework to decide on an approach to determine market country rights to taxation of digital income. I suggest that the measurable factors to be used for determining U.S. source be decided upon following any international consensus that is developed. If no international consensus occurs, then I suggest that measurable factors be limited to those digital service situations where a user's participation or personal content adds value to the cloud service business. Considering the historic focus in regulations and case law on the location where services are performed, merely being a passive user or customer of cloud services should not be sufficient to claim that there is a source of services income from sales to those users/customers. On the other hand, if an MNC or other taxpayer has marketing, sales, and/or customer service personnel in multiple locations, then those personnel (head count, compensation costs, etc.) could be a factor in determining source. (As noted earlier, where any U.S. taxpayer is directly earning cloud services income, it will receive the benefits of tax treaties in effect between the U.S. and the countries from which that taxpayer is earning revenues. Where such a treaty grants taxing rights to the country of the customers/users, there will often be treaty provisions allowing foreign source treatment of that income in the taxpayer's FTC limitation calculation.)

Note that all of the above are objective, tangible, and measurable factors. Subjective factors, such as the value of intangible assets, including both trade and marketing intangibles, should not be used.

In regard to this point about intangible assets and their subjectivity, the following is from a recent article:

The idea that intangible assets are the main creators of value rather than people has ruled supreme for years in the transfer pricing discipline. Although that idea is superficially true, economists are in general agreement that intangibles themselves are the product of cumulative investments in human capital. Irving H. Plotkin and Dan Axelsen provide a thorough synopsis of the economic literature on the role of intangibles in economic growth and the importance of human capital (including education, creative talent, experience, and decision-making ability) in developing intangibles. [Cf. Plotkin and Axelsen, “The Three-Factor Formula vs. the Sources of Income in the New and Weightless Economy,” Tax Mgmt. Int'l J. (Jan. 2013).] The intangibles-centric view often distracted transfer pricing practitioners from the importance of the work entailed to develop, enhance, maintain, protect, and exploit intangible assets (the so-called DEMPE functions) and the location of that work, and (until the BEPS project) it led them to overemphasize the tax jurisdiction of legal ownership of intangibles. [Footnotes omitted. From Ara Stepanyan and Steven D. Felgran, “People Functions Redux: A New Approach to Profit-Splitting Factors”, Tax Notes Federal, September 23, 2019, P. 2035.]

The above point saying “intangibles themselves are the product of cumulative investments in human capital” leads to an important point for the above-noted personnel factor. There has been considerable attention in recent years on the DEMPE functions. Usually in discussion, they are all lumped together with no consideration of their varying natures. Their natures, though, differ importantly when thinking about their relevance to issues of transfer pricing or sourcing of income. In particular, development and enhancement functions can create intangible assets. Maintenance, protection, and exploitation, on the other hand, are involved in conducting a business that uses the developed and enhanced intangible assets.

Considering this distinction as well as Stepanyan and Felgran's point about “cumulative investments in human capital”, it is appropriate in any source-of-income calculation that the development and enhancement personnel costs (including any relevant asset costs) be determined on a cumulative basis or perhaps on a formula basis that includes, say, the current and prior two years or maybe the current year plus 75% of the immediately preceding year, 50% of the second preceding year, 25% of the third preceding year. There's no right or wrong answer on the approach or formula to use. However, the concept is important that a cumulative measurement of relative development and enhancement costs in each country is the best measure for sourcing because it truly accounts for the intangible assets created and avoids the terrible subjectivity of trying to value intangibles and determine their location.

This point about cumulative costs for development and enhancement costs might also apply to personnel (and any relevant asset costs) involved in marketing and sales activities that truly develop marketing intangibles.

In contrast to the above costs that develop intangibles and require some cumulative accounting, maintenance, protection, and exploitation costs are period costs that should not be cumulated.

The above comments about intangibles involve self-developed intangibles. There is a need to deal with intangibles purchased at arm's length from unrelated persons. (Due to the unitary approach, any internal group transfers would be ignored.) Given that there is an actual price paid for such purchased intangible assets, they should be included in the assets factor. As the value of such intangibles should decline with the passage of time and further development and enhancement, the same formula basis as suggested above could be used to reduce the value in future years' source computations. As for location, the location of personnel involved in development and enhancement will of course determine the location of further development and enhancement of any purchased intangibles. The purchased intangibles themselves, though, should be treated as located at the location (or locations) of the personnel who made the commercial/investment decision to acquire the assets.

Personnel, assets, and other factors (e.g. users, sales, etc.) when applied to a taxpayer's income or taxpayer group's unitary income must of course be appropriately weighted. Given the importance of the contribution of personnel both to the development of intangible assets and the actual management and conduct of any business, it seems appropriate to suggest that the personnel factor be given a greater weighting in comparison to other factors. Thus, for example, if there are three factors (e.g., personnel, assets, and user contributions), the personnel factor could be given twice the weighting of the other two factors Thus, the weighting would be 50% for personnel, 25% for assets, and 25% for user contributions. If there were only two factors (e.g. personnel and assets), it would be two-thirds for personnel and one-third for assets.

Simplifying Approaches

As indicated earlier, sourcing under this suggested expansion of Reg §1.861-4 is a facts and circumstances determination. If Treasury and the IRS were to wish to do so, especially given the apparent broad authority under §863(b), they could examine common business models and, through identifying common characteristics, set specific factors and weightings for those common business models. The regulation or other IRS document providing such factors and weightings could specify that taxpayers employing such business models must use the set factors and weightings unless they establish that other factors and weightings are more appropriate.

(For more details about this simplifying approach, see Jeffery M. Kadet, “Expansion of the Profit-Split Method: The Wave of the Future,” Tax Notes Int'l, Mar. 30, 2015, p. 1183. This article includes several examples of concrete allocation keys and weightings as applied to several common business models.)

As a second and more simplified approach, it would not be unreasonable if the Treasury and IRS were to decide to apply a blanket weighting rule for cloud services akin to those applied to certain transportation income under §863(c)(2) or to income from the sale of taxpayer-manufactured property under Reg §1.863-3(b)(1). Given the greater proportion of value that should come from the actual performance of the service in contrast to value provided by users or the country of sale, perhaps an 80%/20% split could be appropriate. (If this simplified approach is considered at all, the split should only be determined after seeing what consensus, if any, is reached by the ongoing OECD Inclusive Framework international dialogue. For example, a consensus that provides for relatively limited market country rights might support, say, a 90%/10% split.)

Once a percentage split is determined for the two factors (i.e., performance of services and users/country of sale), it is necessary to determine the source(s) for each factor.

Performance of services factor. In cloud service businesses, physical assets are of course important. However, their value is typically relatively small in comparison to the potential revenues. Further, there can be some taxpayer latitude in where to place such assets if their location were important to the taxation treatment. Accordingly, it is suggested in this second simplified approach that the 80% factor for performance of services be determined solely on the basis of the location of group personnel.

User/country of sale factor. As for the location of the 20% factor, as above, it is suggested that a decision on this await any consensus reached by the OECD's Inclusive Framework. In the absence of any consensus, then where user contributions add value, the location would be the location of the user. Where users provide no value, then the location would be the location of the customer that is paying for the service. This location could be different, for example, in the case of platforms that generate advertising revenues where advertisers can be located in countries other than where users are located.

Similar to what is described in the section above for a full facts and circumstances approach, both the personnel factor and the user/country of sale factor would be calculated on a group-wide unitary basis and then applied to the cloud service income earned by each group member.

As a brief example to assure clarity, say that X and Y (as described above in “Unitary Application of Reg §§1.861-4 to Cloud Service Businesses”) have a combined 100 relevant personnel with 75 in the U.S. and 25 outside the U.S. Further, their business model does not involve valuable user participation so that only the location of paying users/customers is considered. These paying users/customers in the U.S. generate $50 million of revenue and those outside the U.S. generate $50 million. The “performance of services” factor is 75 U.S./25 foreign, while the “user/country of sale” factor is 50 U.S./50 foreign. With the 80%/20% weighting of the two factors, the income source would be 70% U.S. and 30% foreign (U.S.: 75 x 80% + 50 x 20% = 70%; foreign 25 x 80% + 50 x 20% = 30%).

This second approach, of course, would lack finesse and have less than perfect accuracy. It would, however, allow significant simplification and should provide results that are fair to both the government and taxpayers.

* * * *

If you have any questions concerning the above, I would be pleased to communicate either by email or phone.

Very truly yours,

Jeffery M. Kadet
(206) 395-9849
jeffkadet@gmail.com
Seattle, WA