Respondent determined deficiencies in petitioner’s Federal income and branch profits tax for the taxable years 1988, 1989, and 1990, in the amounts of $518,102, $23,730, and $71,662, respectively.
Unless otherwise indicated, all section references are to the Internal Revenue Code in effect for the taxable years at issue, and all Rule references are to the Tax Court Rules of Practice and Procedure. The sole issue for decision is whether the Convention and Protocols Between the United States and Canada With Respect to Taxes on Income and Capital, Sept. 26, 1980, T.I.A.S. No. 11087 (effective August 16, 1984), 1986-2 C.B. 258 (Canadian Convention), override section 842(b), which requires a foreign company conducting an insurance business in the United States to treat a minimum amount of net investment income as effectively connected with its conduct of that business. For the reasons set forth below, we hold that article VII of the Canadian Convention, 1986-2 C.B. at 260, overrides section 842(b).
FINDINGS OF FACT
Some of the facts have been stipulated and are found accordingly. The stipulation of facts and accompanying exhibits are incorporated herein by this reference. The facts found are those which, unless otherwise specified, existed during the years at issue.
A. Petitioner
The North West Life Assurance Co. of Canada (petitioner) is a life insurance company organized under the corporation laws of Canada with its principal place of business located in Vancouver, British Columbia, Canada. Petitioner operates its life insurance business solely in the United States and Canada and is in the business of writing deferred annuities and life insurance policies. Petitioner began operating in the United States (U.S. branch) in 1971, selecting the State of Washington as its State of entry and subjecting itself to the insurance laws of that State and to regulation by that State’s insurance commissioner. Petitioner maintains a sales and underwriting office in Bellevue, Washington. In addition, petitioner is licensed to transact business as a life insurance company in 20 other States.
Petitioner’s U.S. branch uses a calendar year accounting period and the accrual method of accounting. Petitioner timely filed its Federal income tax returns (Forms 1120L) for tax years 1988, 1989, and 1990.
B. Petitioner’s Product Mix
Petitioner’s U.S. branch operates primarily in the “section 403(b) market”, selling individual deferred annuities to school teachers. Petitioner has the following product mix, measured by its reserves, during the years at issue:
[[Image here]]
Petitioner’s U.S. branch sold these products in the United States, and petitioner’s principal office in Vancouver sold them in Canada.
C. Pricing of Products
Each of petitioner’s annuity contracts includes an accumulation period and a payout period. During the accumulation period, petitioner collects the premiums on its annuity contracts (accumulation annuities). Petitioner does not charge fixed premiums; rather, the annuity holders pay in as much as they desire. Petitioner invests the collected premiums and guarantees its U.S. annuity holders, on a yearly basis, a specific rate of return (1-year rate guaranties). Petitioner makes primarily 5-year interest rate guaranties to its Canadian annuity holders. Petitioner’s annuity holders are able to withdraw the accumulated funds from petitioner once annually during the accumulation period. These withdrawals are subject to surrender charges. The surrender charges are reduced during the first 5 to 10 years of each annuity contract’s existence but are always eliminated before the payout period begins.
During the payout period, petitioner pays the annuity holders fixed periodic payments over the remainder of the annuitant’s life or over a specified number of years (payout annuities). Once the payout period begins, petitioner does not permit early withdrawals.
D. Investment Strategy
Arthur W. Putz, vice president of investments and secretary of petitioner, is primarily responsible for handling the administrative details of petitioner’s investment activity. Donald R. Francis, executive vice president and appointed actuary of petitioner, is primarily responsible for providing actuarial services to petitioner’s life insurance business.
As part of its investment strategy for its U.S. operations, petitioner sought to avoid the risk of fluctuations in currency-exchange and interest rates. Petitioner avoids currency-exchange risk by investing in assets in the same currencies as its insurance liabilities. Petitioner attempts to reduce its interest-rate risk by matching the duration of its assets with the maturity of its liabilities. Washington State law allows an insurance company to invest up to 65 percent of its portfolio in mortgages. Wash. Rev. Code Ann. sec. 48.13.265 (West Supp. 1990). Petitioner invested between 58 percent and 63 percent of its portfolio in mortgages during the years at issue. In order to match its investments in mortgages with the 1-year rate guaranties on its annuities and also enjoy a relatively high return from such investments, petitioner purchases mortgages with 5-year maturities, with a right of renewal for another 5 years at market-adjusted interest rates. The average duration of these mortgages is approximately 2% years. Because petitioner’s 5-year mortgages are longer than the 1-year rate guaranties, part of petitioner’s strategy is to balance its portfolio by also investing in assets with a duration shorter than its liabilities.
Petitioner makes longer term investments in assets backing both its individual life insurance policies and payout annuities than it does in assets supporting its accumulation annuities. Petitioner’s accumulation annuities make up approximately 99 percent of petitioner’s annuity contracts arising from its U.S. branch and approximately 50 percent of the contracts arising from its Canadian office.
E. Flow of Funds
Petitioner collects premiums arising from its U.S. branch business in U.S. currency (U.S. dollars). Upon receipt, for administrative convenience, petitioner transfers the premium payments into a U.S.-dollar-denominated account with Toronto-Dominion Bank in Vancouver, British Columbia (Toronto bank). The Toronto bank pays nominal interest on balances in the account in excess of $250,000.
Washington State law requires foreign insurance companies to maintain a trust account (trusteed assets) in order to qualify to transact insurance in the State. Wash. Rev. Code Ann. sec. 48.05.090 (West Supp. 1990). Petitioner maintains a trust account and an operating account at Seattle First National Bank in Seattle, Washington (Seattle bank). Periodically, petitioner transfers the premiums and interest from the Toronto bank account to the Seattle bank accounts. Petitioner transfers the majority of such funds to the trust account and the balance to the operating account. Petitioner pays commissions, claims, and operating expenses from its operating account. The Seattle bank does not pay any interest on the funds in the operating account. Petitioner invests the premiums in the Seattle trust account in U.S.dollar-denominated assets and retains the earnings in the same account. As a general business practice, during the years at issue, petitioner did not withdraw assets until they matured or rely upon assets outside of the trust account to cover the liabilities incurred by its U.S. branch.
In 1987, petitioner transferred between $7 and $8 million in Canadian-dollar-denominated bonds from its Canadian business to the Seattle bank trust account in order to increase its surplus assets held in the United States relative to the proportion of its surplus held in the Canadian operation. In 1988, petitioner sold stock in a related domestic company for its original cost to Industrial Alliance Life Insurance Co., petitioner’s Canadian parent corporation. The stock had been recorded on the books of petitioner’s U.S. branch and included in the Seattle trust account.
F. Mandatory Filings
The insurance commissioner of each State in which petitioner is licensed to carry on an insurance business requires petitioner’s U.S. branch to file certain annual statements reflecting its U.S. branch operations. To standardize reporting requirements, all States require reporting on the annual statement forms developed by National Association of Insurance Commissioners (NAIC), a voluntary association of State insurance commissioners. NAIC publishes standard detailed forms upon which each type of insurance company reports its annual financial condition.
NAIC form 1A must be filed annually by petitioner with the State of Washington. NAIC form 1A requires information regarding whether a U.S. branch has sufficient admissible assets (all assets of its U.S. branch other than the separate-accounts business) over liabilities, including the statutory deposit. The inside cover of NAIC form 1A states:
This Annual statement differs in some respects from that for a United States Company and should not be interpreted in the same manner. The most important fact conveyed by the statement is whether the Company has a sufficient amount of admissible assets to meet all known liabilities of its United States business including statutory deposit. For this reason, the Annual statement balance sheet does not show the amount of unassigned funds, or surplus, which are accrued from earnings of the United States business, but rather total United States admissible assets and total United States liabilities and statutory deposit.
NAIC form 1 must be filed by domestic insurance companies with their respective State regulatory agencies. Differences between NAIC form 1A and NAIC form 1 include:
1. NAIC form 1A lists assets and liabilities with the assets not necessarily equaling liabilities, capital, and surplus, whereas NAIC form 1 includes a balance sheet;
2. NAIC form 1A lists income and expenses, but it does not include realized capital gains and losses;
3. Schedule D of NAIC form 1A reflects deposits and withdrawals of securities from a trust account at book value, whereas NAIC form 1 reflects purchases and sales of bonds and stocks at transaction prices;
4. NAIC form 1A does not include a reconciliation of capital and surplus from the prior year to the current year, but NAIC form 1 does include such a reconciliation.
The Office of the Superintendent of Financial Institutions Canada (OSFl), Ottawa, Canada, also requires petitioner to file an annual statement (OSFl statement) reflecting its total business in both Canada and the United States. The reporting and accounting requirements for assets, liabilities, income, and expenses for purposes of the OSFl statement are different in a number of respects from those for NAIC forms.
G. Petitioner’s Assets and Surplus
Petitioner reports on its NAIC form 1A the following percentage distribution of assets relating to its U.S. operations:
[[Image here]]
Based on its OSFI statements, petitioner has the following percentage distribution of assets in connection with its worldwide operations:
[[Image here]]
Washington State law requires a foreign insurance company to maintain trusteed assets (equal to the excess of assets over general account liabilities) of at least $2 million. Wash. Rev. Code Ann. sec. 48.05.340(1) (West Supp. 1990). For 1988, 1989, and 1990, petitioner’s NAIC form 1A listed its U.S. branch as having an excess of admissible assets over liabilities in the amounts of $15,422,162, $19,016,749, and $19,363,533, respectively. Petitioner’s ratio of excess mean assets to mean total liabilities are as follows:
[[Image here]]
For each year at issue, a life insurance company incorporated under the laws of the State of Washington would have been in compliance with minimum capital and surplus requirements if it had owned the same assets and incurred the same liabilities as petitioner’s branch, as reported on petitioner’s NAIC form 1A.
H. Computation of Income
During each year at issue, petitioner reported on its Federal income tax returns its net investment income effectively connected with the conduct of its business within the United States, computed pursuant to section 842(a), without regard to the amount of minimum effectively connected net investment income computed pursuant to section 842(b)(1). During the years at issue, petitioner used its NAic.form 1A data to identify to what extent its net investment income was effectively connected for purposes of section 842(a).
Upon audit of petitioner’s Federal tax returns for the years 1988 through 1990, respondent increased petitioner’s income by the extent petitioner’s net investment income computed pursuant to section 842(b) exceeded its income computed pursuant to section 842(a):
[[Image here]]
Respondent did not include an adjustment based on petitioner’s net investment income for 1989. All of the “increases in income tax” for 1988 and 1990 are attributable to the adjustments of petitioner’s taxable income resulting from the application of section 842(b).
I. Treasury Methodology
The Department of the Treasury calculates the asset/liability percentage (i.e., the mean of assets of domestic insurance companies divided by the mean of total insurance liabilities of those same domestic companies) and the domestic investment yield (i.e., the net investment income of domestic insur-anee companies divided by the mean of assets of those same domestic insurance companies) for purposes of section 842(b) using the financial data obtained from the A.M. Best Co. The A.M. Best Co. compiled the data from the naic forms 1 filed by domestic insurance companies with their respective State insurance regulatory authorities. The Treasury considers only data from those companies that appeared in the A.M. Best Co. files for both the second and third years preceding the year at issue (2-year aggregate data). For the years at issue, the Treasury calculated the following asset/liability percentages and domestic investment yields:
[[Image here]]
J. Motion for Entry of Decision
On October 31, 1994, respondent filed a motion for entry of decision. On November 1, 1994, petitioner objected to respondent’s motion. On November 30, 1994, a hearing was held on respondent’s motion. On December 5, 1994, respondent’s motion was denied.
OPINION
I. Statutory Framework
A. Section 842 and Section 864(c)
Under section 842(a),2 a qualified foreign company carrying on a life insurance business within the United States is taxable on its income effectively connected with its conduct of any trade or business within the United States under sub-chapter L, part I. Domestic life insurance companies are also taxed pursuant to the latter provisions. Secs. 801-818. Section 864(c)3 and the regulations thereunder govern when income is effectively connected to petitioner’s business within the United States for purposes of section 842(a). Section 842(b)4 prescribes, by statutory formula, a minimum amount of net investment income that a foreign insurance company, which is taxable under subchapter L, part I, must treat as effectively connected to its conduct of an insurance business in the United States (minimum ECNll). In effect, a foreign insurance company engaged in business in the United States would be taxable, under Internal Revenue Code provisions in issue before us, on the greater of its actual effectively connected net investment income (actual ECNll) pursuant to section 842(a) or its minimum ECNll as determined by the statutory formula.
B. Formula
A foreign insurance company’s minimum ECNII is the product of the company’s required U.S. assets and the domestic investment yield (domestic yield). Sec. 842(b)(1). The required U.S. assets of a company are determined by multiplying the mean of its total insurance liabilities on its business within the United States for the taxable year by the domestic asset/liability percentage (asset/liability percentage) applicable to such company for that year. Sec. 842(b)(2).5 The asset/liability percentage is a ratio, the numerator of which is the mean of the assets of domestic insurance companies and the denominator of which is the mean of the total insurance liabilities of the same domestic insurance companies. Sec. 842(b)(2)(C).6 The domestic yield is a ratio, the numerator of which is the total net investment income of domestic life insurance companies and the denominator of which is the mean annual balance of the total assets of these same domestic companies. Sec. 842(b)(3).7
Section 842(b)(2)(C) and (3) directs the Secretary to calculate both the asset/liability percentage and the domestic yield each year. Section 842(c)(4) provides that each calculation for any taxable year “shall be based on such representative data with respect to domestic insurance companies for the second preceding taxable year as the Secretary considers appropriate.”
C. Worldwide Election
Section 842(b)(4) permits a foreign insurance company to elect to use its own worldwide current investment yield (worldwide yield) rather than the domestic yield.8 A company’s worldwide yield is obtained by dividing the net investment income of the company from all sources by the mean of all assets of the company. Sec. 842(b)(4)(B). A company may not revoke the election without the consent of the Secretary. Sec. 842 (b)(4)(C).
II. Canadian Convention
The Canadian Convention is designed to prevent double taxation and to avoid fiscal evasion (Preamble to Canadian Convention). Article vil of the Canadian Convention governs when and how much of the profits of a qualified Canadian enterprise is subject to U.S. Federal income tax. The relevant provisions of article VII for making such a determination are as follows:
1. The business profits of a resident of a Contracting State shall be taxable only in that State unless the resident carries on business in the other Contracting State through a permanent establishment situated therein. If the resident carries on, or has carried on, business as aforesaid, the business profits of the resident may be taxed in the other State but only so much of them as is attributable to that permanent establishment.
2. Subject to the provisions of paragraph 3, where a resident of a Contracting State carries on business in the other Contracting State through a permanent establishment situated therein, there shall in each Contracting State be attributed to that permanent establishment the business profits which it might be expected to make if it were a distinct and separate person engaged in the same or similar activities under the same or similar conditions and dealing wholly independently with the resident and with any other person related to the resident * * *
3. In determining the business profits of a permanent establishment, there shall be allowed as deductions expenses which are incurred for the purposes of the permanent establishment, including executive and general administrative expenses so incurred, whether in the State in which the permanent establishment is situated or elsewhere. Nothing in this paragraph shall require a Contracting State to allow the deduction of any expenditure which, by reason of its nature, is not generally allowed as a deduction under the taxation laws of that State.
s{; íJí # ‡
5. For the purposes of the preceding paragraphs, the business profits to be attributed to a permanent establishment shall be determined by the same method year by year unless there is good and sufficient reason to the contrary.
ijc ‡ ijc if: ifc
7. For the purposes of the Convention, the business profits attributable to a permanent establishment shall include only those profits derived from the assets or activities of the permanent establishment.
[Canadian Convention, art. VII, 1986-2 C.B. at 260; emphasis added.]
Article XXV, paragraph (6) of the Canadian Convention states in pertinent part:
6. Notwithstanding the provisions of Article XXIV (Elimination of Double Taxation), the taxation on a permanent establishment which a resident of a Contracting State has in the other Contracting State shall not be less favorably levied in the other State than the taxation levied on residents of the other State carrying on the same activities. * * * [Canadian Convention, art. XXV, par. (6), 1986-2 C.B. at 268.]
In the instant case, the parties agree that petitioner is entitled to the benefits of the Canadian Convention and that petitioner operates its insurance business in the United States through a U.S. permanent establishment.
Congress can override a convention provision by enacting a subsequent statute. Reid v. Covert, 354 U.S. 1, 18 (1957). Congress ratified the Canadian Convention in 1984. Canadian Convention, Sept. 26, 1980, T.I.A.S. No. 11087, 1986-2 C.B. 258 (effective Aug. 16, 1984). It initially appears that Congress sought to override the Canadian Convention in the Omnibus Budget Reconciliation Act of 1987, Pub. L. 100-203, sec. 10242(a), 101 Stat. 1330, 1330-420, by amending section 842 to incorporate subsection (b). In the conference report to section 842(b), Congress stated, however, that it did “not intend to apply the general principle that, in the case of a conflict, a later enacted statute prevails over earlier enacted statutes or treaties.” H. Conf. Rept. 100-495, at 983 (1987), 1987-3 C.B. 193, 263.
Respondent contends that we should construe the Canadian Convention so as to harmonize the convention with the statute. If, however, we find that the Canadian Convention and section 842(b) conflict, respondent concedes that the convention prevails and that no deficiencies in income tax or branch profits tax for the years at issue exist.
Petitioner does not challenge the taxation of its actual ecnii or respondent’s calculations of its minimum ECNII for any of the years at issue. Accordingly, if we find that the Canadian Convention and section 842(b) are consistent, petitioner concedes that section 842(b) applies in this case and that it owes the income and branch profits tax as determined by respondent in her statutory notices of deficiency.
The parties present various alternative arguments based on provisions of article vil and article XXV. In deciding whether petitioner is entitled to relief from section 842(b) as a result of the Canadian Convention, we must determine whether:
1. Section 842(b), in requiring petitioner to report a minimum amount of ECNII, conflicts with the requirements of paragraphs (1), (2), and (7) of article VII on how to determine the profits attributable to a permanent establishment;
2. section 842(b) violates paragraph (5) of article VII, which requires a consistent method of profit attribution to be applied unless a good and sufficient reason to the contrary exists; or
3. section 842(b) violates article xxv, paragraph (6) by levying taxation less favorably on petitioner than the Internal Revenue Code levies taxation on U.S. residents carrying on the same activities.
We discuss these issues in the context of the relevant convention articles. The issues before us are of first impression.
III. Principles of Convention Obligations
Before addressing the parties’ arguments pertaining to specific convention provisions, we consider the principles for interpreting conventions.
The goal of convention interpretation is to “give the specific words of a * * * [convention] a meaning consistent with the genuine shared expectations of the contracting parties”. Maximov v. United States, 299 F.2d 565, 568 (2d Cir. 1962), affd. 373 U.S. 49 (1963). Courts liberally construe treaties to give effect to their purpose. United States v. Stuart, 489 U.S. 353, 368 (1989); Bacardi Corp. of Am. v. Domenech, 311 U.S. 150, 163 (1940). Even where a provision of a treaty fairly admits of two constructions, one restricting, the other enlarging, rights which may be claimed under it, the more liberal interpretation is to be preferred. United States v. Stuart, supra at 368. In construing a convention, we give the language its ordinary meaning in the context of the convention, unless a more restricted sense is clearly intended. De Geofroy v. Riggs, 133 U.S. 258, 271 (1890). Finally, it is well settled that when a convention and a statute relate to the same subject, courts will always attempt to construe them so as to give effect to both. Estate of Burghardt v. Commissioner, 80 T.C. 705, 713 (1983), affd. without published opinion 734 F.2d 3 (3d Cir. 1984). “Although not conclusive, the meaning attributed to treaty provisions by the Government agencies charged with their negotiation and enforcement is given great weight”. United States v. Stuart, supra at 369 (citing Kolovrat v. Oregon, 366 U.S. 187, 194 (1961)).
The Model Double Taxation Convention on Income and on Capital, Report of the O.E.C.D. Committee on Fiscal Affairs (1977) (Model Treaty), and explanatory commentaries (Model Commentaries) provide helpful guidance. See Letter of Transmittal from President Carter to the Senate of the United States requesting ratification of the Convention, dated November 12, 1980, 4 Roberts & Holland, Legislative History of United States Tax Conventions 242 (1986); S. Comm. on Foreign Relations, Tax Convention and Proposed Protocols with Canada, S. Exec. Rept. 98-22 (1984), 4 Roberts & Holland, supra at 1096; Taisei Fire & Marine Ins. Co. v. Commissioner, 104 T.C. 535 (1995) (use of O.E.C.D. Commentaries in interpreting meaning of permanent establishments). It is the role of the judiciary to interpret international conventions and to enforce domestic rights arising from them. See Kolovrat v. Oregon, 366 U.S. 187 (1961); Perkins v. Elg, 307 U.S. 325 (1939); Charlton v. Kelly, 229 U.S. 447 (1913); United States v. Rauscher, 119 U.S. 407 (1886). Tax treaties are purposive, and, accordingly, we should consider the perceived underlying intent or purpose of the treaty provision. See, e.g., Estate of Burghardt v. Commissioner, supra at 717 (treating a reference to a “specific exemption” in a U.S.-Italy estate tax treaty as not limited to an exemption as such, but included a subsequently enacted unified credit having the same function as an exemption); Smith, “Tax Treaty Interpretation by the Judiciary”, 49 Tax Law. 845, 858-867 (1996). In addressing the issues of this case, we shall keep at the forefront our role in the interpretation of conventions.
Respondent asserts two principles of convention interpretation with which petitioner disagrees. First, respondent argues that the literal terms of a convention must be interpreted consistently with the expectations and intentions of the United States in entering the Canadian Convention. In support of this contention, respondent cites United States v. Stuart, supra at 365-366, and Sumitomo Shoji Am., Inc. v. Avagliano, 457 U.S. 176, 185 (1982). Second, respondent represents that the principles of treaty interpretation, set forth in her brief, were approved by the Office of International Tax Counsel of the Treasury Department as interpretations consistently held by the United States. Respondent contends that any contrary interpretations held by Canada are subordinate to such consistently maintained U.S. interpretations. Respondent relies upon United States v. A.L. Burbank & Co., 525 F.2d 9 (2d Cir. 1975) for support of her contention.
None of these cases supports respondent’s position. As evidenced by Sumitomo Shoji Am., Inc. v. Avagliano, supra at 180, and later in United States v. Stuart, 489 U.S. at. 365-366, the Supreme Court has consistently held that we must consider the expectations and intentions of both signatories, not just those of the United States. The Court states:
The clear import of treaty language controls unless “application of the words of the treaty according to their obvious meaning effects a result inconsistent with the intent or expectations of its signatories.” [United States v. Stuart, supra at 365-366 (citing Sumitomo Shoji Am., Inc. v. Avagliano, supra at 180 (quoting Maximov v. United States, 373 U.S. at 54)).]
Moreover, we do not agree with respondent’s contention that A.L. Burbank & Co. stands for the proposition that the Government’s position is entitled to deference at the expense of our convention partner’s interpretation. In A.L. Burbank & Co., the Canadian tax authorities requested the Internal Revenue Service (the Service) to obtain information to assist them in their Canadian tax investigation. The Canadian authorities made their request pursuant to the 1942 tax convention between the United States and Canada. Convention on Double Taxation, Mar. 4, 1942, U.S.-Can., T.S. No. 983, 56 Stat. 1399. The United States had no interest in the investigation, and there was no claim that U.S. income taxes were due. The Service’s understanding of the Canadian position was that Canadian tax authorities might not act on a reciprocal request to obtain information for the United States unless Canadian taxes were also at issue. The Court of Appeals for the Second Circuit held that even if Canada failed to satisfy its reciprocal obligation under the convention, the United States was permitted to use the summons authority of section 7602 to obtain the information requested by Canadian tax officials. United States v. A.L. Burbank & Co., supra at 15.
As we stated above, our goal is to construe the Canadian Convention according to the “genuine shared expectations of the contracting parties”. Maximov v. United States, 299 F.2d at 568. While the meaning attributed to treaty provisions by Government agencies charged with their negotiation and enforcement can be very helpful to us, and we give great weight to that meaning, United States v. Stuart, supra at 369, deference is not the same as blind acceptance. See Coplin v. United States, 6 Cl. Ct. 115 (1984), revd. on other grounds 761 F.2d 688 (Fed. Cir. 1985), affd. 479 U.S. 27 (1986). There is no authority for the proposition that a court construing a convention must follow the interpretation suggested by our Government when that interpretation runs contrary to what the court concludes was the intent of the contracting parties. Id. Indeed, the Supreme Court has noted that “courts interpret treaties for themselves,” Kolovrat v. Oregon, 366 U.S. at 194, and that the construction given by Government agencies is not conclusive. Sumitomo Shoji Am., Inc. v. Avagliano, supra at 184. The deference afforded depends upon the degree to which the interpretation proffered by respondent, as the official U.S. position, is reasonable, unbiased, and consistent with what appear to be -the circumstances surrounding the convention. Coplin v. United States, supra. As discussed below, other evidence in the record undermines the plausibility of respondent’s position and hence the deference that the Court is able to afford to that interpretation.
IV. Article VII of the Canadian Convention
Petitioner argues that paragraphs (1), (2), and (7) of article vil of the Canadian Convention require that profits be attributed to its permanent establishment as if the latter were a separate entity distinct from petitioner’s head office, with income measured by reference to the permanent establishment’s own specific operations. Petitioner goes on to argue that the statute mandates the application of section 842(b) in all instances where there is effectively connected investment income. If the actual income is less than the minimum under the statute, then the provision applies — a result that, in petitioner’s opinion, conflicts with article VII, paragraphs (1), (2), and (7), which petitioner interprets to preclude taxing Canadian companies on a fictional amount that is greater than their actual income derived from their business in the United States.
Respondent raises various arguments supporting why section 842(b) is consistent with article VII of the Canadian Convention and contends: (1) Section 842(b) is a permissible method of attributing profits to a permanent establishment under article vil; (2) section 842(b) serves as a backstop to section 842(a) and corrects any underreporting by foreign insurance companies of their actual ECNll; and (3) the U.S. Senate, which advised and consented to the Canadian Convention and approved section 842(b), believed that section 842(b) was consistent with the Canadian Convention.
In our view, resolution of this controversy depends on the interpretation given to article VII, paragraphs (2) and (5). While article VII, paragraph (1) limits U.S. taxation of income earned by a Canadian enterprise to the income “attributable” to the enterprise’s permanent establishment, article vil, paragraphs (2) and (5) direct how those attributable profits are to be determined. Article VII, paragraph (2) limits “attributable” profits to those which a “distinct and separate person engaged in, the same or similar activities under the same or similar conditions” would be expected to make (hereafter referred to as the separate-entity principle or basis). Article VII, paragraph (5) requires that profits be attributed by the same method each year unless there is a good and sufficient reason to the contrary. To satisfy the convention obligations of the United States, the domestic rules of attribution must determine the profits attributable to petitioner’s permanent establishment within the limits set forth therein. Our analysis begins by considering how to measure the profits on a separate-entity basis and whether section 842(b) determines minimum amounts of ECNII in a manner consistent with those limits.
V. Measurement of Profits on a Separate-Entity Basis
Petitioner argues that the language of article vil requires income to be attributed to a permanent establishment based on its own> particular operations. Respondent argues that article VII does not require a specific method or guarantee mathematical certainty and that, consequently, either country may use its domestic law in determining the profits attributable to a permanent establishment.
It is axiomatic that the “Interpretation of the * * * Treaty * * * must, of course, begin with the language of the Treaty itself.” Sumitomo Shoji Am., Inc. v. Avagliano, 457 U.S. at 180. As we stated above, the clear import of treaty language controls. Id. But article vil, paragraph (2) speaks in ambiguous terms, and when language is susceptible to differing interpretations, extrinsic materials bearing on the parties’ intent should be considered. Day v. Trans World Airlines, Inc., 528 F.2d 31, 34 (2d Cir. 1975); Hidalgo County Water Control & Improvement District v. Hedrick, 226 F.2d 1, 8 (5th Cir. 1955).
The Senate’s preratification materials confirm that the Canadian Convention was based in part on the Model Treaty. See S. Exec. Rept. 98-22, supra at 3. Our examination shows that the business profits article of the Model Treaty9 includes provisions substantially similar to article VII, paragraphs (1) and (2) of the Canadian Convention. While the Model Treaty itself provides no more explanation than the Canadian Convention on how to determine the profits attributable to a permanent establishment, the Model Treaty is explained in part by the Model Commentaries. Petitioner relies upon paragraphs 10 and 13 of the Model Commentaries to article 7, paragraph (2) of the Model Treaty in support of its contention that the separate-entity language of article VII, paragraph (2) requires that taxable profits be attributed to a permanent establishment based on the establishment’s facts. These paragraphs provide in pertinent part:
10. This paragraph contains the central directive on which the allocation of profits to a permanent establishment is intended to be based. The paragraph incorporates the view, which is generally contained in bilateral conventions, that the profits to be attributed to a permanent establishment are those which that permanent establishment would have made if, instead of dealing with its head office, it had been dealing with an entirely separate enterprise under conditions and at prices prevailing in the ordinary market. Normally, these would he the same profits that one would expect to be determined by the ordinary processes of good business accountancy. * * *
13. Clearly many special problems of this kind may arise in individual cases but the general rule should always be that the profits attributed to a permanent establishment should be based on that establishment’s accounts insofar as accounts are available which represent the real facts of the situation. * * *
[Model Commentaries to article 7, paragraph (2) of the Model Treaty; emphasis added.]
In her trial memorandum, respondent acknowledges: “[The model] [Commentaries] [express] a preference for an arm’s-length standard for the ‘distinct and separate person’ entity with separate accounts”. Respondent contends, however, that article VII permits either country to apply its domestic law in determining the profits attributable to a permanent establishment. In this regard, respondent relies upon the Technical Explanation, prepared by the Treasury Department and submitted to the Senate Foreign Relations Committee. The Technical Explanation states in pertinent part:
Paragraph 7 provides a definition for the term “attributable to”. Profits “attributable to” a permanent establishment are those derived from the assets or activities of the permanent establishment. Paragraph 7 does not preclude Canada or the United States from using appropriate domestic tax law rules of attribution. The “attributable to” definition does not, for example, preclude a taxpayer from using the rules of section 1.864 — 4(c)(5) of the Treasury Regulations to assure for U.S. tax purposes that interest arising in the United States is attributable to a permanent establishment in the United States. (Interest arising outside the United States is attributable to a permanent establishment in the United States based on the principles of Regulations sections 1.864-5 and 1.8G4-6 and Revenue Ruling 75-253, 1975-2 C.B. 203.) Income that would be taxable under the Code and that is “attributable to” a permanent establishment under paragraph 7 is taxable pursuant to article VII, however, even if such income might under the Code be treated as fixed or determinable annual or periodical gains or income not effectively connected with the conduct of a trade or business within the United States. The “attributable to” definition means that the limited “force-of-attraction” rule of Code section 864(c)(3) does not apply for U.S. tax purposes under the Convention. [Technical Explanation by the Treasury Department of the Convention Between the United States of America and Canada with Respect to Taxes on Income and on Capital Signed at Washington, D.C. on September 26, 1980, as Amended by the Protocols Signed on June 14, 1983 and March 28, 1984, 4 Roberts & Holland, Legislative History of United States Tax Conventions, at 1020, 1032 (1986); 1986-2 C.B. 275, 279.]
In the alternative, respondent infers from the absence of any reference in the Technical Explanation to a conflict between the Canadian Convention and prior section 819(a) of the Internal Revenue Code of 195410 that Canada implicitly accepted that attribution rules such -as section 842(b) would apply.
Nevertheless, we are satisfied that petitioner’s construction of the separate-entity principle of article vil, paragraph (2) is correct. The extrinsic evidence and the Model Tréaty and Commentaries, on which the Canadian Convention is based in part, support that construction. The Senate’s preratification materials to the Canadian Convention do not ascribe a different meaning to the separate-entity language of article vil, paragraph (2). See S. Exec. Rept. 98-22, at 20 (1984). Moreover, it is consistent with the approach historically taken by the United States and Canada. Art. m(l) of the Convention on Double Taxation, Mar. 4, 1942, U.S.-Can., T.S. No. 983, 56 Stat. 1399.11
While the Treasury’s interpretation, set forth in the Technical Explanation, is particularly persuasive in 'light of the fact that the Canadian Department of Finance has generally accepted the Technical Explanation as an accurate portrayal of the understandings and context in which the Canadian Convention was negotiated, see ALI Project, 18 (1992); Canadian Department of Finance, Rel. No. 81-6 (Feb. 4, 1981), we think that respondent misconstrues the Treasury’s interpretation. We are not persuaded that the language set forth in the Technical Explanation of article VII, paragraph (7), see supra p. 384, was intended to interpret article VII as preserving the right to the use of all of the domestic attribution rules. In this context, the Technical Explanation’s use of the word “paragraph” takes on a significant meaning. The word “paragraph” in the Technical Explanation’s discussion of article VII, paragraph (7) signals that under paragraph (7) of article VII, domestic rules of attribution may remain viable. It does not necessarily follow that all domestic rules remain so under the rest of article VII, particularly article VII, paragraph (2). To adopt respondent’s interpretation would require us to substitute the word “article” for “paragraph” and would render the limit imposed by the separate-entity principle meaningless. By way of contrast, our interpretation of the Technical Explanation, as it relates to article Vil, paragraph (7), gives effect to the word “paragraph” without modification but still preserves the rest of article VII.
In a similar vein, we decline to infer an implicit acceptance of attribution rules like section 842(b), on the part of Canada, from the fact that the Technical Explanation does not mention any conflict between the Canadian Convention and section 819(a). As we discuss below, see infra pp. 387-391, there is a superficial similarity between prior section 819(a) and section 842(b), but, nevertheless, there are important differences between them.
Accordingly, we hold that the disposition of this case turns on whether the section 842(b)(1) formula prescribes a minimum amount of ECNII based on the facts' as they relate to petitioner’s permanent establishment, by reference to the establishment’s separate accounts insofar as those accounts represent the facts of the situation, and by the same method each year unless there is a good and sufficient reason to do otherwise. It is to that review that we direct our attention.
Petitioner retained Dale S. Hagstrom and Daniel J. McCarthy of Milliman & Robertson, Inc.12 (Hagstrom), whose report endeavored to analyze the hypothetical impact of applying the section 842(b) formula to the domestic insurance industry, and/or to U.S. branches of Canadian insurance companies. Without going into the details of the conclusions reached by petitioner’s experts, suffice it to say that, overall, we do not find their analysis to be helpful. For example, significantly, section 842(b) does not apply to domestic insurance companies. Furthermore, we agree with petitioner that section 842(b) attributes a fictional amount of income to petitioner’s U.S. branch that is not based on its own activities but rather on the investment performance achieved by domestic insurance companies.
Respondent contends that section 842(b) does not violate the separate-entity principle because the formula therein uses petitioner’s liabilities to determine the assets petitioner might be expected to hold if it were a separate entity. Whether the hypothetical amount of assets calculated pursuant to section 842(b) represents a reasonable estimate of the amount of assets petitioner would hold if it were a separate entity misses the point; that amount is simply extraneous to petitioner’s operations. Section 842(b) incorporates domestic insurance industry data via the domestic yield or company’s worldwide earnings data via the worldwide yield, all of which are extraneous to the operations of petitioner’s U.S. permanent establishment. We are not persuaded that the separate-entity principle is satisfied merely by starting with the real facts as they relate to petitioner’s permanent establishment but then incorporating extraneous data that is inconsistent with that principle. Cf. Ostime (Inspector of Taxes) v. Australian Mut. Provident Socy., [1960] AC 459 (United Kingdom case with a similar business profits article stating that “the worldwide investment income, which forms the first stage of the * * * calculation of profits, cannot be attributed to the hypothetical independent enterprise without violating the very hypothesis which * * * the treaty is designed to lay down as the basis of taxability”; i.e., the separate-entity principle). Respondent’s own witness, Dr. Newlon, an international economist with the Treasury, admitted that the formula could be improved. We are convinced that section 842(b) is contrary to and inconsistent with article VII, paragraph (2), which precludes the fictional allocation of business profits to petitioner’s, permanent establishment.
Imputing a level of assets and yields to petitioner’s U.S. branch, respondent contends, is not unreasonable because the formula incorporates actual business data and petitioner operates in the U.S. market and directly competes with domestic life insurance companies. To conclude that section 842(b) is reasonable in light of the fact that petitioner operates in the United States would not resolve the dispute before us. It is not enough for section 842(b) to be reasonable. To sustain the application of section 842(b) based on the facts before us, we must conclude that it comports with our Canadian Convention obligation. See United States v. A.L. Burbank & Co., 525 F.2d at 15. As we have stated above, we must conclude that the statute does not; consequently, it cannot prevail in the presence of the Canadian Convention.
Respondent asserts that article VII, paragraph (2) permits the use of formulas in determining the taxable profits under limited circumstances. In support, respondent relies upon paragraph 23 of the Model Commentaries to article 7, paragraph (3) of the Model Treaty, which states in pertinent part:
23. It is usually found that there are, or there can be constructed, adequate accounts for each part or section of an enterprise so that profits and expenses, adjusted as may be necessary, can be allocated to a particular part of the enterprise with a considerable degrée of precision. This method of allocation is, it is thought, to be preferred in general wherever it is reasonably practicable to adopt it. There are, however, circumstances in which this may not be the case and paragraphs 2 and 3 are in no way intended to imply that other methods cannot properly be adopted where appropriate in order to arrive at the profits of a permanent establishment on a “separate enterprise” footing. It may well be, for example, that profits of insurance enterprises can most conveniently be ascertained by special methods of computation, e.g. by applying appropriate co-efficients to gross premiums received from policy holders in the country concerned. Again, in the case of a relatively small enterprise operating on both sides of .the border between two countries, there may be no proper accounts for the permanent establishment nor means of constructing them. There may, too, be other cases where the affairs of the permanent establishment are so closely bound up with those of the head office that it would be impossible to disentangle them on any strict basis of branch accounts. Where it has been customary in such cases to estimate the arm’s length profit of a permanent establishment by reference to suitable criteria, it may well be reasonable that that method should continue to be followed, notwithstanding that the estimate thus made may not achieve as high a degree of accurate measurement of the profit as adequate accounts. Even where such a course has not been customary, it may, exceptionally, be necessary for practical reasons to estimate the arm’s length profits.
Respondent argues that paragraph 23 of the Model Commentaries permits the adoption of formulas if any one of the following circumstances is satisfied: (1) Formulas are found to be more convenient or administratively necessary; (2) the permanent establishment lacks adequate accounts by which to determine the attributable profits; (3) the other method is customary; (4) the permanent establishment is a foreign insurance enterprise; and (5) an exceptional need for the method is demonstrated. In respondent’s view, each circumstance is satisfied in the instant case, and section 842(b) is a permissible method by which to determine attributable profits within the meaning of the Canadian Convention. The amicus curiae brief submitted by the Government of Canada asserts that when contracting parties to a tax convention intend to permit the use of formulas to determine profits of a permanent establishment engaged in the insurance business, the convention will contain a specific provision to that effect.
We need not decide whether article VII, paragraph (2) permits the use of formulas in determining the profits attributable to a permanent establishment. As a preliminary matter, we find respondent’s reliance on paragraph' 23 of the Model Commentaries to be misplaced. We think respondent gives paragraph 23 too broad a reading. Our reading leads us to the conclusion that, at a minimum, before other methods (other than using the accounts of a permanent establishment) may be adopted under the guidance of paragraph 23, the other method must be customary and based on suitable criteria or the circumstances must be exceptional.
Respondent contends that section 842(b) is customary because it is substantially similar to the prior sections 819(a)13 (applying from years 1959 through 1983) and 813 of the Internal Revenue Code of 1954, as amended, and section 813 of the Internal Revenue Code of 198614 (applying from years 1984 through 1987).
The prior sections 819(a) and 813 both required a foreign insurance company to compare its branch’s actual surplus (excess of assets over total insurance liabilities) held in the United States to a statutory minimum surplus. Secs. 819(a)(2)(A), 813(a)(1). If the branch’s surplus was less than a statutory minimum, the deficiency was treated as additional assets of the branch which were deemed to have earned the same yield that the branch had earned on the assets it actually held. Secs. 819(a)(1), 805(b)(2), 813(a)(3). The Deficit Reduction Act of 1984, Pub. L. 98-369, sec. 211(a), 98 Stat. 720, 743, repealed prior section 819 and added prior section 813. The prior section 819(a) required a foreign life insurance company to reduce certain deductions by the product of the deficiency and the foreign insurance company’s actual yield. Sec. 819(a). The Deficit Reduction Act of 1984 modified how taxable income was calculated for foreign insurance companies. Sec. 801; see H. Rept. 98-861, at 1043-1044 (1984), 1984-3 C.B. (Vol. 2) 1, 297-298. This change necessitated treating the product of the deficiency and the foreign insurance company’s actual yield as additional effectively connected income instead of as a reduction of certain deductions which was done under the previous section 819.
Subsequently, Omnibus Budget Reconciliation Act of 1987, Pub. L. 100-203, sec. 10242, 101 Stat. 1330, 1330-420, repealed prior section 813 and added section 842(b). We do not find that section 842(b) is similar enough to prior sections 819 and 813 so as to establish that section 842(b) was customary within the meaning of paragraph 23 of the Model Commentaries. Two features of section 842(b) go beyond the historical approach taken by both of the earlier statutes. First, section 842(b)(1) imputes additional income based on an earnings yield derived from domestic industry averages, sec. 842(b)(3), or petitioner’s worldwide operations, sec. 842(b)(4), whereas both prior section 819 and section 813 imputed income based on the U.S. branch’s actual earnings yield, secs. 819(a)(1)(B), 805(b)(2), 813(a)(3). Second, section 842(b) applies an entirely new yield to all of the branch’s assets, not just to the additional imputed assets. Prior sections 819(a) and 813, on the other hand, imputed additional income for just those deemed assets.
We recognize that a convention, like a constitution, is a dynamic instrument, drafted to take account of changing conditions and expectations. See Day v. Trans World Airlines, Inc., 528 F.2d at 35; Maximov v. United States, 299 F.2d at 568. If we were to accept respondent’s argument, however, the United States could, through various amendments to the Internal Revenue Code, always eliminate unilaterally the separate-entity principles in article VII, paragraph (2) without ever violating the Canadian Convention.
Nor are we persuaded that the circumstances herein are exceptional. While paragraph 23 does not prescribe when circumstances are considered exceptional, we do have other guidance as to when such circumstances may exist. Paragraph 11 of the Model Commentaries states that
11. In the great majority of cases, trading accounts of the permanent establishment * * * will be used by the taxation authorities concerned to ascertain the profit properly attributable to that establishment. Exceptionally there may be no separate accounts (cf. paragraphs 23 to 27 below). * * * [Emphasis added.]
The above language suggests that other methods may only be adopted when a permanent establishment does not have any accounts. As we discuss below, the real facts (accounts) are ascertainable in the instant case.
Respondent also seeks to justify the application of section 842(b) on the grounds that the statute serves as a necessary backstop to section 842(a) and corrects any underreporting of actual ECNII by foreign insurance companies. The parties agree that petitioner as well as other foreign insurance companies use their NAIC form 1A data to identify to what extent their net investment income is effectively connected to their U.S. businesses for purposes of section 842(a).
In this context, respondent contends that foreign insurance companies have significant discretion in moving their assets between taxing jurisdictions once their State statutory trust requirements are satisfied. Respondent points out that Washington State law does not require a foreign insurer to deposit income earned on trusteed assets in the trust account and that it permits petitioner to replace trusteed assets with other assets of equal value and quality, subject to the investment rules. Consequently, NAIC forms 1A, respondent argues, fail to reflect the economic realities of the businesses of foreign insurance companies operating in the United States and are unreliable for the purpose of determining their actual ECNII. Respondent goes on to argue that absent section 842(b), foreign insurance companies may escape U.S. taxation on investment income attributable to their U.S. business.
Respondent retained Richard E. Stewart, Richard S.L. Roddis, and Barbara D. Stewart of Stewart Economics, Inc.15 (Stewart), as expert witnesses to give their professional opinion as to the reasons and incentives Canadian life insurance companies have for holding certain assets in the United States and why the investment income earned on those assets is not an inherently reliable measure of the investment income flowing from the branch operations. We have received into evidence their report.16 Stewart reviewed the history of the State regulations applying to foreign insurance companies, including the NAIC annual statement and the trust requirements for such companies. The report agrees with respondent that NAIC form 1A is not an effective means by which to ascertain the net investment income that is effectively connected to an insurance business within the United States and that NAIC form 1A (foreign insurers’ form) differs significantly from NAIC form 1 (domestic insurers’ form). Stewart concluded that
Canadian life insurance companies are not required to earmark specific assets for their U.S. business * * *. [B]ecause Canadian life insurance companies have economic incentives to place higher yielding assets in lqwer taxing jurisdictions, something other than NAIC statement assets and investment yields are needed to determine the investment income derived from the trusteed assets of a Canadian life insurance company.
Respondent also points to various facts that she claims indicate that petitioner’s NAIC forms 1A fail to reflect the economic realities of petitioner’s U.S. branch: (1) During 1988 through 1990, petitioner’s total cash and term deposits, which were maintained as a part of its U.S. branch, were 75 percent, 90 percent, and 75 percent, respectively, of its total worldwide funds; (2) petitioner maintained only 35 percent, 38 percent, and 50 percent of its total bond portfolio— arguably higher yielding assets — in its U.S. branch, for 1988, 1989, and 1990, respectively; (3) petitioner transferred Canadian-dollar-denominated bonds from its Canadian business to its Seattle bank trust account in order to equalize the surplus held in each operation; and (4) petitioner transferred from its Seattle bank trust account to its Canadian parent stock that petitioner held in a subsidiary, and for purposes of the transfer the stock was valued at its cost rather than at its fair market value.
We agree with respondent and respondent’s expert that the NAIC form 1A is not the ideal means for reconciling and identifying all of the income attributable to a permanent establishment. It does not include a closed, self-contained book of accounts, reconciliation of any surplus, or information regarding capital gains or losses. The form is not designed to identify taxable income but rather to monitor compliance with State regulatory requirements on trusteed assets. That conclusion, however, does not resolve the issue before us.
The record is clear that petitioner occasionally exercised its discretion in moving assets between jurisdictions as evidenced by petitioner’s transfer of its Canadian bonds from its Canadian operations to its Seattle bank trust account and by the sale of its stock in a subsidiary to its foreign parent. Through the testimony of Mr. Putz and Mr. Francis, whom we found to be informative and credible witnesses, petitioner has established, however, as a general business practice, that it did not commingle assets between its Seattle bank trust account and its Canadian investment portfolio and had separate investment strategies in each country.
We are satisfied with their explanations of the business reasons behind petitioner’s investment strategy. According to Mr. Francis’ testimony, petitioner’s U.S. branch had significant liquidity demands as a result of its need to balance its 5-year mortgage holdings. Petitioner has established that, in fact, it avoided currency risk and only invested assets in the same currencies as its insurance liabilities because of the narrow profit margins on its products.
As petitioner correctly points out, if petitioner’s accounts were considered so inherently unreliable as to justify ignoring those accounts for purposes of the Canadian Convention, such a method should be used in all years, not just when the statute produces a higher amount than do petitioner’s accounts. Article VII, paragraph (5) of the Canadian Convention makes clear that the same method of profit allocation is to be used each year unless there is a “good and sufficient reason to the contrary.” Paragraph 30 of the Model Commentaries to article 7, paragraph (6)17 explains that “a method of allocation once used should not be changed merely because in a particular year some other method produces more favourable results”. The parties stipulated that for 1989 petitioner’s actual ECNII and minimum ECNII were $19,910,031 and $19,606,065, respectively. As a result, section 842(b) would not increase petitioner’s net investment income for 1989 because petitioner’s actual ECNII exceeded petitioner’s minimum ECNII. On the other hand, those accounts that were adequate for 1989 were deemed inadequate for years 1988 and 1990 solely on the basis that petitioner’s minimum ECNII of $21,282,045 and $20,749,629 exceeded its actual ECNII of $18,501,669 and $20,426,754 and not based on the actual inaccuracies of those accounts. We find that section 842(b) contravenes the basic premise set forth in article VII, paragraph (5) of the Canadian Convention.
In the totality of petitioner’s circumstances, we do not believe that petitioner underreported its actual ECNII during the years at issue despite whatever deficiencies may exist in using NAIC form 1A to identify the extent to which petitioner’s net investment income was effectively connected.
Section 842(b) has the effect of penalizing petitioner, who reported income commensurate with its U.S. business but whose investment performance does not attain the U.S. average in each year. Such an approach is simply not consistent with either article VII, paragraph (2) or (5).
Respondent argues that petitioner’s facts are not representative of the foreign insurance industry in the United States. Respondent admits that petitioner may be adversely affected by section 842(b) as written but contends, as a policy matter, the Court should not find the statute to be inconsistent with the Canadian Convention merely because one particular taxpayer is adversely affected if the Court concludes that the statute as a whole is designed to achieve the appropriate results for taxpayers in general over the long term. But both parties agree that this case does not turn on the validity of the policy reasons underlying the adoption of section 842(b) (i.e., the potentially abusive ability of foreign insurance companies to hold excess liquid assets outside of the United States or to hold higher yielding assets outside of the United States), and we agree with them.
Respondent retained Christian DesRochers of the Avon Consulting Group18 to give his professional opinion as to the economic impact of section 842(b). DesRochers analyzed section 842(b) and the hypothetical impact of applying the formula therein to the U.S. branches of Canadian insurance companies.
DesRocher’s analysis of the impact of section 842(b) on taxpayers not before us is of little help. As a threshold matter, respondent’s argument raises an issue as to the proper factual focus of our review. Respondent argues that the United States, Canada, and petitioner all accepted that a domestic attribution rule would be tested against article vu on the basis of the circumstances of all Canadian life insurance companies rather than just on a particular taxpayer’s facts. To support this contention, respondent relies in part on the discussion in the Technical Explanation. See supra p. 384. In respondent’s view, because both countries expected all of their respective domestic attribution rules to apply, it follows that each country expected the domestic rule to be reviewed based on the facts of the entire industry.
We need not engage in a detailed analysis of whether various foreign insurance companies pay Federal income tax in accordance with our tax laws.19 Respondent’s argument essentially ignores the language of the referenced discussion of the Technical Explanation. See supra p. 384. As we previously discussed, see supra pp. 384-385, we do not believe that the United States and Canada intended for all domestic attribution rules to be preserved under article VII.
Throughout article VII and particularly article VII, paragraph (2), the language therein refers only to a single permanent establishment rather than the industry in which the establishment operates. When the language is reasonably clear, as it is in this particular context, the party proffering a contrary interpretation must persuade the court that its construction comports with the view of both parties. See Sumitomo Shoji Am., Inc. v. Avagliano, 457 U.S. at 180.
In light of the foregoing, the language and purpose of article vil, paragraph (2) and the content of the Canadian Convention as a whole, we also do not believe that the approach suggested by respondent could have been within the “shared expectations of the contracting parties,” Maximov v. United States, 299 F.2d at 568, and, consequently, we do not agree that petitioner implicitly accepted respondent’s approach to interpreting article VII.
Respondent is generally correct that section 842(b) was intended to serve as a backstop to the rules in section 842(a) and section 864(c). The conference report to section 842(b) states:
The conferees understand that the provision governing foreign insurance companies solves a statutory problem in the context of the broader issue: measuring the U.S. taxable income of a foreign corporation that is effectively connected with its U.S. trade or business. That issue more generally involves the determination of which of the corporation’s assets generate gross effectively connected income, and which of its expenses and liabilities are connected with such income. Certain types of assets and liabilities that must, in this process, be attributed in whole or in part to a U.S. trade or business may be particularly suitable for movement among various trades or businesses of a single foreign corporation, may be fungible with assets and liabilities identified with other trades or businesses of the corporation, or may be usable by more than one such trade or business simultaneously. * * * [H. Conf. Rept. 100-495, at 984 (1987), 1987-3 C.B. 193, 264.]
But such a conclusion does not affect the outcome of this case. To begin with, as previously noted, the issues in this case do not concern policy in section 842(b) or any other provisions of the Internal Revenue Code. Rather, the issues concern whether the statute comports with our convention obligations. Unfortunately, in the instant case, section 842(b) cannot survive in the presence of the Canadian Convention.
Finally, respondent argues that we should construe section 842(b) as agreeing with the Canadian Convention because the U.S. Senate, which advised and consented to the Canadian Convention and approved the statute, believed the statute did not violate any existing conventions. In support, respondent points to several statements in the conference report to section 842(b).20
We are not persuaded by respondent’s assertion that this statement in the conference report should guide the result in this case. To the extent that the statements in the conference report may be read as expressing the view of the Senate that section 842(b) is consistent with the Canadian Convention, they are the statements of a subsequent Senate and, therefore, at best, “form a hazardous basis for inferring the intent of an earlier one.” South Carolina v. Regan, 465 U.S. 367, 379 n.16 (1984); Consumer Prod. Safety Commn. v. GTE Sylvania, 447 U.S. 102, 117 (1980). In the end, the courts alone must declare what the Canadian Convention and particularly article Vil mean. See American Exch. Sec. Corp. v. Helvering, 74 F.2d 213, 214 (1934).
In sum, we are confronted with a situation in which the language of article vn, paragraph (2) is at best murky, and the interpretations of both parties have advantages and disadvantages. We are impressed that the Canadian Convention may give an economic advantage to Canadian insurance companies operating through a permanent establishment in the United States. Nevertheless, our view is that petitioner’s interpretation of article VII, paragraph (2) best carries out the intent of the United States and Canada as set forth in the Canadian Convention and satisfies the purpose of article VII of the Canadian Convention — to attribute income to a permanent establishment based on its real facts, and, accordingly, we so hold.
Having found that petitioner is entitled to relief from section 842(b) based on article VII, paragraph (2) of the Canadian Convention, we have no need to delve into the question of whether petitioner is also entitled to such relief based on article XXV, paragraph (6).
Finally, we note that respondent did not contend that section 482 applied in the instant case. Accordingly, our decision in the instant case does not consider the application of section 482 in those circumstances in which the Canadian Convention also applies.
The Executive Branch with the advice and consent of the Senate has the option of negotiating a new protocol with Canada creating an exception similar to one included in subsequent conventions. These conventions contain a general directive to determine profits as if the taxpayer was a separate entity, yet also include explicit exceptions permitting each country to apply its own internal methods of taxation to the business profits of an insurance company’s permanent establishment. See, e.g., art. 7(7), Tax Convention, July 23, 1982, U.S.-N.Z., 35 U.S.T. (Part 2) 1949, 1990-2 C.B. 274.21 We have considered all of the other arguments made by respondent and, to the extent we have not addressed them, find them to be without merit.
Decision will be entered for petitioner.
Reviewed by the Court.
Cohen, Chabot, Jacobs, Gerber, Parr, Wells, Whalen, Beghe, Laro, and Vasquez, JJ., agree with this majority opinion. Chiechi, J., did not participate in the consideration of this opinion.Sec. 842(a) provides in pertinent part:
SEC. 842(a). Taxation Under This Subchapter. — If a foreign company carrying on an insurance business within the United States would qualify under part I + * * of this subchapter for the taxable year if (without regard to income not effectively connected with the conduct of any trade or business within the United States) it were a domestic corporation, such company shall be taxable under such part on its income effectively connected with its conduct of any trade or business within the United States * *
Sec. 864(c) provides in pertinent part:
(2) Periodical, etc., income from sources within United States — factors.—In determining whether income from sources within the United States of the types described in section 871(a)(1), section 871(h), section 881(a), or section 881(c) or whether gain or loss from sources within the United States from the sale or exchange of capital assets, is effectively connected with the conduct of a trade or business within the United States, the factors taken into account shall include whether—
(A) the income, gain, or loss is derived from assets used in or held for use in the conduct of such trade or business, or
(B) the activities of such trade or business were a material factor in the realization of the income, gain, or loss.
In determining whether an asset is used in or held for use in the conduct of such trade or business or whether the activities of such trade or business were a material factor in realizing an item of income, gain, or loss, due regard shall be given to whether or not such asset or such income, gain, or loss was accounted for through such trade or business.
*******
(4) Income from sources without United States.—
* * * * * * *
(C) In the case of a foreign corporation taxable under part I * * * of subchapter L, any income from sources without the United States which is attributable to its United States business shall be treated as effectively connected with the conduct of a trade or business within the United States.
Sec. 842(b) provides in pertinent part:
(1) In GENERAL. — In the case of a foreign company taxable under part I * * * of this sub-chapter for the taxable year, its net investment income for such year which is effectively connected with the conduct of an insurance business within the United States shall be not less than the product of—
(A) the required U.S. assets of such company, and
(B) the domestic investment yield applicable to such company for such year.
Sec. 842(b)(5) defines net investment income for purposes of sec. 842(b) as follows:
Net INVESTMENT INCOME. — For purposes of this subsection, the term “net investment income” means—
(A) gross investment income (within the meaning of section 834(b)), reduced by
(B) expenses allocable to such income.
Sec. 842(b)(2) provides:
(2) Required U.S. assets.—
(A) In GENERAL. — For purposes of paragraph (1), the required U.S. assets of any foreign company for any taxable year is an amount equal to the product of—
(i) the mean of such foreign company’s total insurance liabilities on United States business, and
(ii) the domestic asset/liability percentage applicable to such foreign company for such year.
(B) Total INSURANCE liabilities. — For purposes of this paragraph—
(i) COMPANIES TAXABLE UNDER part I. — In the case of a company taxable under part I, the term “total insurance liabilities” means the sum of the total reserves (as defined in section 816(c)) plus (to the extent not included in total reserves) the items referred to dn paragraphs (3), (4), (5), and (6) of section 807(c).
Sec. 842(b)(2)(C) provides in pertinent part:
(C) Domestic asset/liability percentage. — The domestic asset/liability percentage applicable for purposes of subparagraph (A)(ii) to any foreign company for any taxable year is a percentage determined by the Secretary on the basis of a ratio—
(i) the numerator of which is the mean of the assets of domestic insurance companies taxable under the same part of this subchapter as such foreign company, and
(ii) the denominator of which is the mean of the total insurance liabilities of the same companies.
Sec. 842(b)(3) provides:
(3) Domestic investment yield. — The domestic investment yield applicable for purposes of paragraph (1)(B) to any foreign company for any taxable year is the percentage determined by the Secretary on the basis of a ratio—
(A) the numerator of which is the net investment income of domestic insurance companies taxable under the same part of this subchapter as such foreign company, and
(B) the denominator of which is the mean of the assets of the same companies.
Sec. 842(b)(4) provides in pertinent part:
(4) Election to use worldwide yield.—
(A) In GENERAL. — If the foreign company makes an election under this paragraph, such company’s worldwide current investment yield shall be taken into account in lieu of the domestic investment yield for purposes of paragraph (1)(B).
(B) Worldwide current investment yield. — For purposes of subparagraph (A), the term "worldwide current investment yield” means the percentage obtained by dividing—
(i) the net investment income of the company from all sources, by
(ii) the mean of all assets of the company (whether or not held in the United States).
(C) Election. — An election under this paragraph shall apply to the taxable year for which made and all subsequent taxable years unless revoked with the consent of the Secretary.
Art. 7 of Model Double Taxation Convention on Income and on Capital, Report of the O.E.C.D. Comm, on Fiscal Affairs (1977) (Model Treaty) provides in pertinent part:
Par. 1. The profits of an enterprise of a Contracting State shall be taxable only in that State unless the enterprise carries on business in the other Contracting State through a permanent establishment situated therein. If the enterprise carries on business as aforesaid, the profits of the enterprise may be taxed in the other State but only so much of them as is attributable to that permanent establishment.
Par. 2. Subject to the provisions of paragraph 3, where an enterprise of a Contracting State carries on business in the other Contracting State through a permanent establishment situated therein, there shall in each Contracting State be attributed to that permanent establishment the profits which it might be expected to make if it were a distinct and separate enterprise engaged in the same or similar activities under the same or similar conditions * * *
Congress enacted sec. 819(a) as part of the Life Insurance Company Income Tax Act of 1959, Pub. L. 86-69, sec. 2, 73 Stat. 136.
Art. III(l) in the second income tax convention with Canada signed in 1942 provided in pertinent part:
1. If an enterprise of one of the contracting States has a permanent establishment in the other State, there shall be attributed to such permanent establishment the net industrial and commercial profit which it might be expected to derive if it were an independent enterprise engaged in the same or similar activities under the same or similar conditions. Such net profit will, in principle, be determined on the basis of the separate accounts pertaining to such establishment. * * *
Mr. Hagstrom holds a B.A. in mathematics frjm Princeton University. He is a Fellow of the Society of Actuaries and a member of the American Academy of Actuaries. Mr. McCarthy holds a B.S. in mathematics from Fordham University. In addition, he is a Fellow of the Society of Actuaries and a charter member of the American Academy of Actuaries. He has been designated as an enrolled actuary by the Joint Board for the Enrollment of Actuaries.
Sec.- 819(a) of the Internal Revenue Code of 1954, as amended and in effect for 1983, provided in pertinent part:
(1) In general. — In the case of any foreign corporation taxable under this part, if the minimum figure determined under paragraph (2) exceeds the surplus held in the United States, then—
(A) the amount of the policy and other contract liability requirements (determined under section 805 without regard to this subsection), and
(B) the amount of the required interest (determined under section 809(a)(2) without regard to this subsection),
shall each be reduced by an amount determined by multiplying such excess by the current earnings rate (as defined in section 805(b)(2)).
(2) Definitions. — For purposes of paragraph (1)—
(A) The minimum figure is the amount determined by multiplying the taxpayer’s total insurance liabilities on United States business by a percentage for the taxable year to be determined and proclaimed by the Secretary.
The percentage determined and proclaimed by the Secretary under the preceding sentence shall be based on such data with respect to domestic life insurance companies for the preceding taxable year as the Secretary considers representative. Such percentage shall be computed on the basis of a ratio the numerator of which is the excess of the assets over the total insurance liabilities, and the denominator of which is the total insurance liabilities.
Sec. 805(a) of the 1954 Internal Revenue Code, as amended, defined policy and other contract liability requirements as the sum of:
(1) the adjusted life insurance reserves, multiplied by the adjusted reserves rate,
(2) the mean of the pension plan reserves at the beginning and end of the taxable year, multiplied by the current earnings rate, and
(3) the interest paid.
Sec. 805(b)(2) of the 1954 Internal Revenue Code, as amended, defined the current .earnings rate as:
* * * the amount determined by dividing—
(A) the taxpayer’s investment yield for such taxable year, by
(B) the mean of the taxpayer’s assets at the beginning and end of the taxable year.
Sec. 805(b)(4) defined assets for purposes of the above sec. 805(b)(2) as follows "all assets of the company (including nonadmitted assets), other than real and personal property (excluding money) used by it in carrying on an insurance trade or business.”
Sec. 809(a)(2) of the 1954 Internal Revenue Code, as amended, defined “required interest” for purposes of subsection 819 as:
the sum of the products obtained by multiplying—
(A) each rate of interest required, or assumed by the taxpayer, in calculating the reserves described in section 810(c) by
(B) the means of the amount of such reserves computed at that rate at the beginning and end of the taxable year.
The Life Insurance Company Income Tax Act of 1959, Pub. L. 86-69, sec. 2, 73 Stat. 136, added sec. 819(b), which required the same adjustment as sec. 819(a) except that the minimum figure was determined by multiplying the foreign life insurance company’s total insurance liabilities on U.S. business by 9 percent for tax years beginning before January 1, 1959 and by an annual percentage determined by the Treasury for tax years thereafter. The Foreign Investors Act of 1966, Pub. L. 89-809, sec. 104, 104(i)(3), 80 Stat. 1539, 1561, redesignated the adjustment provision as sec. 819(a) for tax years beginning after 1966. From 1966 until 1983, sec. 819(a) remained unchanged except for minor changes, which are not relevant to the instant case.
Sec. 813 provided in pertinent part:
SEC. 813(a). Adjustment Where Surplus Held In the United States Is Less Than Specified Minimum.—
(1) In GENERAL. — In the case of any foreign company taxable under this part, if—
(A) the required surplus determined under paragraph (2), exceeds
(B) the surplus held in the United States, then its income effectively connected with the conduct of an insurance business within the United States shall be increased by an amount determined by multiplying such excess by such company’s current investment yield. * * *
(2) Required surplus. — For purposes of this subsection—
(A) In GENERAL. — The term “required surplus” means the amount determined by multiplying the taxpayer’s total insurance liabilities on United States business by a percentage for the taxable year determined and proclaimed by the Secretary under subparagraph (B).
(B) Determination of percentage. — The percentage determined and proclaimed by the Secretary under this subparagraph shall be based on such data with respect to domestic life insurance companies for the preceding taxable year as the Secretary considers representative. Such percentage shall be computed on the basis of a ratio the numerator of which is the excess of the assets over the total insurance liabilities, and the denominator of which is the total insurance liabilities.
(3) Current investment yield. — For purposes of this subsection—
(A) In general. — The term'“current investment yield” means the percent obtained by dividing—
(i) the net investment income on assets held in the United States, by
(ii) the mean of the assets held in the United States during the taxable year.
Richard E. Stewart holds a degree from West Virginia University and a law degree from Harvard Law School. He is former Superintendent of Insurance of the State of New York and a former officer and director of the Chubb Group of Insurance Companies. Richard S.L. Roddis holds a degree from San Diego University and a law degree from Boalt Hall School of Law at the University of California at Berkeley. He is a former Superintendent of Insurance of the State of California. Barbara D. Stewart holds a bachelor’s degree in economics and business administration from Beaver College. She is a former corporate economist of the Chubb Group.
We have disregarded the Stewart report to the extent that it relies on State law of Washington for years not at issue. The record does not indicate that petitioner operated in Michigan; we have also disregarded the report to the extent that it relies on Michigan State law.
Art. 7(6) of the Model Treaty is substantially similar to Art. VII(5) of the Canadian Convention. Art. 7(6) states in pertinent part: “For the purposes of the preceding paragraphs, the profits to be attributed to the permanent establishment shall be determined by the same method year by year unless there is good and sufficient reason to the contrary.”
Mr. DesRochers holds an undergraduate degree in political science from the University of Connecticut. He is a Fellow of the Society of Actuaries and a member of the American Academy of Actuaries.
Respondent’s proposed findings of fact include data relating to other Canadian life insurance companies carrying on insurance businesses through branches in the United States. On brief, petitioner objected to the relevancy of these findings unless the Court concluded that they were relevant “in light of the ‘test case’ aspect of the proceeding”. Because we have decided the controverted issues without considering these stipulations in our Findings of Fact and our Opinion, the admissibility of these stipulations has become moot. Although we reviewed all the material submitted in this case, we only address facts affecting petitioner.
The conference report, H. Conf. Rept. 100-495, at 983-984 (1944), 1987-3 C.B. 193, 263-264, listed several factors, originally developed by the Treasury indicating why sec. 842(b) and U.S. treaties were consistent: (1) Sec. 842(b) applies to life insurance companies in a manner substantially similar to the present-law rules which the Treasury did not consider to violate U.S. treaties; (2) sec. 842(b) attributes to the U.S. trade or business of a foreign life insurance company an amount of assets determined by reference to the assets of comparable domestic insurance companies, thereby reasonably measuring the amount of assets that the U.S. trade or business of a foreign insurance company would be expected to have were it a separate company dealing independently with non-U.S. offices of the foreign insurance company; and (3) sec. 842(b) furnishes regulatory authority for the Secretary to provide a relief mechanism to mitigate the effects of any increase in tax resulting from the fact that a taxpayer’s deemed income from required U.S. assets exceeds its actual income from those assets.
Art. 7(7) of the U.S.-New Zealand Convention includes a provision regarding the taxation of permanent establishments of insurance companies. This provision provides:
Nothing in this Article shall prevent either Contracting State from taxing according to its law the income or profits from the business of any form of insurance. [Tax Convention, July 23, 1982. U.S.-N.Z.. 35 U.S.T. (Part 2) 1949, 1964.1