Magog (City of) v. Canada, 2001 FCA 210

                                                                                                                                            Date: 20010621

                                                                                                                                         Docket: A-829-99

Coram:             DÉCARY

LÉTOURNEAU

NOËL, JJ.A.

Between:

VILLE DE MAGOG

Appellant

and

HER MAJESTY THE QUEEN

Respondent

JUDGMENT

The appeal is allowed, the decision of the Tax Court of Canada is set aside, and the assessment issued on March 18, 1996 is referred to the Minister of Revenue for re-examination and reassessment on the understanding that the method used by the appellant to compute the claimed repayment is consistent with the requirements of subsection 141.01(5) of the Act and that the appellant is entitled to perform the computation authorized by subsection 21(2) of the Regulations for the 1991, 1993 and 1994 periods, with costs in favour of the appellant at trial and on appeal.

           "Robert Décary"

                                                          J.A.

Certified true translation

Suzanne M. Gauthier, LL.L., Trad. a.

Date: 20010621

                                              Docket: A-829-99

Neutral Citation: FCA 210

Coram:             DÉCARY

LÉTOURNEAU

NOËL, JJ.A.

Between:

VILLE DE MAGOG

Appellant

and

HER MAJESTY THE QUEEN

Respondent

REASONS FOR JUDGMENT

NOËL J.A.

[1]         This is an appeal from a judgment of the Tax Court of Canada (reported at [1999] T.C.J. no. 806 (T.C.C.) (Q.L.)) rejecting the method adopted by the appellant to compute the input tax credit (ITC) to which it was entitled under the Excise Tax Act, S.C. 1994, c. 9 (the "Act").

[2]         The appellant is a municipality that elected under Part V of the Streamlined Accounting (GST) Regulations (the "Regulations") to determine its net tax in accordance with that part. Under subsection 21(2) of the Regulations, a municipality which, in addition to providing the usual services of a municipality, operates an electricity service, shall make a separate calculation for that service in computing its net tax in the manner specified in that subsection. The appellant is such a municipality. The subsection in question reads as follows:

[3]         Moreover, subsection 141.01(5) of the Act stipulates:

[4]         For each of the periods at issue, the appellant computed its net tax in accordance with subsection 21(2) of the Regulations and, to this end, developed its own method for determining the extent to which it consumed or used properties or services for the purpose of providing its electricity service as a taxable supply. It is only in regard to properties and services used for this purpose that it is entitled to the ITC. The issue in this litigation is whether the elected method is fair and reasonable within the meaning of subsection 141.01(5).

[5]         The appellant provides some services that are for strictly municipal purposes. These include, for example, such services as assessment, transportation, recreation, urban planning, waterworks, etc. Other services are mixed, in that they include some municipal service components in the strict sense as well as some components of the electricity service supplied by the municipality.

[6]         In this regard, the appellant's director of finance and treasurer stated in the course of his testimony before the Tax Court of Canada that more than 50% of the town's income is derived from the sale of electricity and that in many instances the supply of this service is incorporated in other services provided by the municipality. He explained, for example, that the Municipal Council acts as the Board of Directors of the electricity service business and is responsible for the regulation, fee structure, payroll policies, etc. Moreover, the director of finance, in addition to his so-called "municipal" duties, is responsible for the invoicing of the 70,000 subscribers to the electricity service as well as the receipts, and it is the Municipal Court that is used for recovery of unpaid bills.

[7]         The so-called "mixed" activities were identified according to the budget items as they appear in the appellant's books for the period in dispute under the titles Legislation, Law Enforcement, Financial and Administrative Management, Record Office, Personnel Management, Building Maintenance and Communications (Appeal Book, vol. I, pages 48 to 54). It is these mixed activities, and in particular the allocation difficulties they present, that are at the heart of this litigation.

[8]         The Act stipulates that the method used to allocate properties acquired in the context of a mixed activity must be fair and reasonable. No particular method is recommended, but the Minister of Revenue has published some guidelines in this regard (Memorandum 700-5-1) addressed in particular to the computation of the ITC by financial institutions.

[9]         Relying on this document, the appellant used a formula based on the cost of the properties and services acquired in the context of its activities (the inputs) to determine the ITC to which it was entitled. Using this method, the appellant determined that it was entitled to an ITC of $34,198.70 for the period in dispute and asked that this amount be refunded. A refund was allowed by way of assessment issued on March 18, 1996, but only to the extent of $2,920.50. The appellant objected to this assessment, without success, and its subsequent appeal to the Tax Court of Canada was likewise dismissed.

[10]       The method used by the appellant to compute the ITC is performed in two stages. First, the appellant isolated the expenses exclusively pertaining to the electricity service and the expenses exclusively pertaining to the municipal services. It then allocated the expenses pertaining to mixed activities in the proportion reflected by these two figures. Thus, for 1994, the expenses exclusively connected with the electricity service amounted to $10,374,046, while those exclusively connected with municipal services amounted to $10,005,170, and it is the percentage of the first figure over the sum of the two, or 50.90%, that was applied to the expenses incurred in the mixed activities for the purpose of identifying the portion of these expenditures that was related to the electricity service.

[11]       The auditor for the Minister of Revenue said he, too, referred to Memorandum 700-5-1 in developing his method. He explained that according to this memorandum mixed activities should if possible be directly allocated and that the input-based method is allowed only to the extent that a direct allocation is not possible. In this case, the auditor explained, he had obtained his results through a partially direct allocation.

[12]       Essentially, the auditor elected to allocate the expenses pertaining to the Legislation, Record Office and Personnel Management items according to the percentage of the employees who were assigned to the electricity service and those assigned to municipal services. For example, for the 1994 period, 9.28% of the employees were assigned to the electricity service and it is this percentage that was used to identify the mixed activities pertaining to the electricity service. In regard to the Financial and Administrative Management item, the auditor applied the appellant's method. The expenses incurred under Law Enforcement were allocated according to an approximate count of bailiff's costs and, finally, the expenses pertaining to Building Maintenance and Communications were allocated according to the estimated use per floor of City Hall, in the first case, and according to the use of pages in the town's monthly publication, in the second (testimony of auditor, Appeal Book, vol. II, pages 275 to 282).

[13]       The difference between the results obtained by the appellant and those obtained by the respondent in regard to the proportion of mixed activities related to the electricity service is:

[14]       The Tax Court of Canada judge concluded that the method adopted by the appellant produced distorted results that contradicted its own data, while the method proposed by the auditor was arrived at in a painstaking fashion in accordance with logical, objective criteria (reasons, paragraph 26). She added, at the end of her analysis:

[This method] accurately reflects the financial situation of the electricity service and, in that sense, is the fair and reasonable method required by subsection 141.01(5) of the Act. [Emphasis added]

Analysis and decision

[15]       The only issue before the judge was whether the method elected by the appellant was fair and reasonable, as required by subsection 141.01(5). She did not have to determine which of the two methods in question was the best. Moreover, Memorandum 700-5-1 acknowledges in its 23^rd paragraph that more than one method may be fair and reasonable within the meaning of the Act (see also Navaho Inn v. The Queen, 3 GTC 2067, at page 2071 (T.C.C.)).

[16]       The fact that the appellant was engaged in some mixed activities during the period at issue is not disputed. The respondent acknowledges that the exempted services provided by the appellant are in many respects rendered together with the electricity service, and that some method had to be used to determine the ITC that is properly attributable to the taxable activity.

[17]       It is important in this regard to note that the Act does not require the appellant to establish the type of accounting systems that would enable it to separate out each property or service that is consumed or used in the context of its mixed activities. Parliament was aware that such a requirement could result in compliance expenses that would exceed the tax yielded. So it left it to the taxpayer to select an appropriate method, while requiring that the method chosen be "fair and reasonable".

[18]       The method chosen by the appellant separates the mixed activities by pro-rating the expenses pertaining exclusively to the electricity service and those pertaining exclusively to the municipal services. The reasoning underlying this method is that the mixed expenses are incurred in the same proportion as the expenses incurred exclusively for either of these purposes. This logic is prima facie sound and seems to lead to a fair and reasonable allocation. Moreover, both the auditor and the judge acknowledged this, as it is this method that was adopted by the auditor in segregating the expenses under the heading Financial and Administrative Management.

[19]       There are two findings of fact underlying the judge's conclusion that the appellant's method was not fair and reasonable despite its apparent logic. She summarizes them as follows, in paragraph 22 of her reasons:

[22] As regards personnel management, the percentage used by the auditor seems, on the face of it, more reasonable than that used by the appellant. As to the profitability of the power system, the inclusion of the large amount of expenses for the mixed activities flagrantly contradicts the appellant's accounting analysis.

In my opinion, the first of these conclusions, assuming it to be correct, is inconsequential in the context of this litigation, and the second is based on a misassessment of the evidence.

[20]       In regard to the first, it will be recalled that the auditor allocated the expenses under the Personnel Management item in proportion to the number of employees assigned to the electricity service and those assigned to the municipal services. The judge found, in regard to this item, that the percentage used was more reasonable than the one used by the appellant. As I said earlier, it was not the judge's role to determine whether one method was more reasonable than the other; rather, she was to determine whether the method used by the appellant was fair and reasonable.

[21]       In any event, the appellant noted that even if the auditor's allocation was more accurate than the one it used, under the Personnel Management item, very few expenses are listed there (Applicant's Record Book, vol. I, page 52), so the judge's conclusion in this regard has no practical impact. In fact, the total ITC claimed by the appellant under the Personnel Management item amounts to $667.59. This is no reason to justify the rejection of the method adopted by the appellant.

[22]       The second reason given by the judge for rejecting the appellant's method is the finding she made at paragraph 18 of her reasons:

Exhibit I-3 is the statement of revenue and expenditure for the electricity service. This is a budget document prepared by the appellant itself. It may be seen therein that, for 1994, total revenue was $13,850,699 and total expenses before apportionment and allocation were $11,614,031. The apportionment of $798,681 is described as the portion assigned to the municipality for energy consumption. The net allocations amount to $8,649. The document does not explain what this expression means. It may be taken to mean the allocation of the expenses of certain mixed activities. The excess of revenue over net expenditures, after apportionment and allocation, is $3,030,700. According to the appellant's reasoning, $5,000,000 should be added to expenditures, as shown in Exhibit A-3 (referred to above in paragraph 14 of these Reasons), thus creating a deficit of $2,000,000. [Emphasis added]

[23]       It is this conclusion that led the judge to state that "the inclusion of the large amount of expenses for the mixed activities flagrantly contradicts the appellant's accounting analysis" (paragraph 22) and that "the method proposed by the appellant produces distorted results which contradict its own data" (paragraph 26).

[24]       However, according to the evidence, there is no basis for the computation made by the trial judge. She wrongly concludes that the revenues from the electricity service ($3,030,700) should be reduced by $5,000,000, "as shown in Exhibit A-3".

[25]       Neither in exhibit A-3 nor in the record as a whole can we find any justification for this sum of $5,000,000. The judge seems to have improperly assumed that the expenses exclusive to the power system (which amount to about $10,000,000) were mixed expenses to which the percentage of 50.90% should have been applied, according to the appellant's method (exhibit 1-3, Appeal Book, vol. I, page 120). In fact, the mixed expenses pertaining to the electricity service amounted to $2,462,848 (Appeal Book, vol. I, page 54).

[26]       Thus, it was through a misassessment of the evidence that the judge drew the conclusion that the appellant's method created a deficit of $2,000,000 for the electricity service. In fact, profits of $1,777,110.40 were generated using this method (Appellant's supplementary memorandum, paragraphs 4 to 18). These profits would have been within 25% of the figure identified by the auditor, using his own method.

[27]       This error is no doubt at the origin of the trial judge's rejection of the appellant's method. In fact, a review of the record discloses that the auditor's method is no better than that of the appellant. It is instructive in this regard to quote two lengthy extracts from the cross-examination of the auditor (Appeal Book, vol. II, at pages 302 and 306, respectively):

[Translation]

Q.     Mr. Lacoste, I come back to your work sheet. You began your work sheet with the FT-7. If I correctly understand the exercise you did, to determine the percentage of the inputs related to the Legislation, the Record Office and the Personnel [Management], you took a relationship between all the municipal employees, who are not working on electricity, ninety-three point five (93.5) and those working on electricity, thirteen point five (13.5).

A.     That is correct.

Q.     To end up saying: fine, I have one hundred and fifty-one point six (151.6) employees in municipal services, there are fifteen (15) in electricity, so the municipal services represent ninety point seven two per cent (90.72%) of all expenses pertaining to the Legislation, Record Office and Personnel Management items.

A.     That is correct.

Q.     This means, for example, the mayor's salary is included in that item.

A.    That is correct.

Q.     So, you concluded that the mayor, through that exercise, spends only nine point two eight per cent (9.28%) of his time managing the hydro-electric system.

A.    That is correct.

A.     To go into it in detail, yes, perhaps I should have done that, but as I had explained to Mr. Gagnon, I said... the law, the memorandums all refer to the most fair and reasonable method, I had told Mr. Gagnon: "We can spend three weeks, one month here going over point by point exactly what it is that each person is doing or each expenditure, what it is used for, or we can try to agree on a simple method and arrive at a percentage," and that is what was done.

In the case of personnel management, it was obvious, the figures are there to demonstrate it; Legislation, and Record Office, a bit as I said earlier, we reached... it was the last two items, what method could we take? We agreed on that basis.

[...]

Q.     But it is my impression that this is a survey method that you used in the case of the legislation, which is to say... rather than take a percentage of inputs, you took a percentage of human resources, saying: the total personnel of the Town represents that, so I am going to apply that ratio to the Record Office, the Legislation and the Personnel Management. Is that correct?

A.     That is correct. But always with the agreement of Mr. Gagnon.

Q.     I understand, but my question is... you use the word "direct method"; for me, the direct "allocation" method is the one to which you referred a minute ago, which would have taken three weeks, the one where -- like the memorandum we will see later -- the one where I could have sat down with the mayor... to ask Mr. Gagnon to put it in writing for me: do you want me to say how much time the mayor, over the four years, spent on management? How much time do you spend per day on management? How much time does the receptionist at the desk have to spend taking telephone calls in relation to that?

A.     Yes.

Q.     And that is what is suggested in the example in the memorandum, to take for each individual... and example A in the memorandum -- and I will come back to this later -- tells us: ten per cent (10%) of the time of this person, it is that; fifteen per cent (15%) of the time of someone else. In your case you used instead a survey, you used a rule of three in a survey context. Nine point two eight per cent (9.28%) of the Town's employees are direct employees in electricity, consequently nine point two eight per cent (9.28%) of the Legislation, Record Office and Personnel Management expenses will be expenses related to hydro-electricity.

A.    In the case of Personnel Management, yes, that is correct.

Q.     Legislation, Record Office and Personnel Management.

A.     Yes. Legislation and Record Office, it was one method, o.k., somewhat arbitrary that we used because in fact the question of time is important there. The department's time is important, namely... I mean, we are there to do an audit file, we want to do it in the most realistic way possible but we have the time constraint also, and Mr. Gagnon has the same thing. I don't think he would have wanted me to spend three weeks down there disturbing all his employees in order to determine the exact percentage.

Q.     I certainly don't want to suggest that to you, Sir, and... all I am telling you is that I am trying to see... because you characterize the method you used in the case of the Legislation as a direct method and I submit to you that it was instead a survey method identical or comparable to the one used by Mr. Leroux, but that you concluded was in your case the best one.

A. I said "direct" in the case of Personnel Management. In the case of Legislation, afterwards I came back and I said: "Fine, those are some methods we used afterward to finalize, basically, the last two points that remained." Another reason, I want to say, there was the time that was important there and the other important thing was the amounts of tax involved. If you look just at Legislation for 94, you have about thirty-five thousand dollars ($35,000) of expenses that may be taxable. But if you compute the taxes there and you compute perhaps an ITC versus the rebate, which is fifty-seven point one four percent (57.14%), well, you calculate it and you say: yes, are you going to spend three weeks just to get a thousand dollars ($1,000).

Anyway, from the Department's standpoint, no, it is not worthwhile, and from Mr. Gagnon's standpoint, I don't think he would have wanted me to justify three weeks of time disturbing his employees in order to determine which was the better method.

[28]       Clearly, then, the auditor too proceeded by way of a survey in regard to the Legislation, Record Office and Personnel Management items, and his method is not any more objective or logical than that of the appellant. The same applies to the Law Enforcement, Building Maintenance and Communications items (Examination of auditor, Appeal Book, vol. I, pages 275 to 282). The only item in regard to which the auditor's method may seem more logical is that of Personnel Management, but nothing flows from that given the low level of expenses inventoried in that item. Considering the fact that both methods at issue involve the use of surveys, the results they produce are not basically incompatible.

[29]       I conclude, therefore, that the judge could not reject the appellant's method on the basis of the auditor's alternative computation, and that the appellant's method is also fair and reasonable within the meaning of subsection 141.01(5).

[30]       To succeed in its appeal, the appellant had to show not only that its method was sound but also, for each of the periods in question, that more than 50% of the property and services subject to this method were acquired in relation to the electricity business. Subsection 21(2) of the Regulations provides, in effect, that the calculation contemplated therein is available only in regard to property and services "primarily" acquired for that purpose (Navaho Inn, supra at page 21-7, Mid-west Feed Ltd. v. M.N.R., [1987] 2 D.T.C. 394 at page 400 (T.C.C.)). Using its own method the 50% threshold was not met by the appellant for the 1992 period.

[31]       For these reasons, I would allow the appeal, I would set aside the decision of the Tax Court of Canada and I would refer the assessment issued on March 18, 1996 to the Minister of Revenue for re-examination and reassessment on the basis that the method used by the appellant to compute the claimed refund is consistent with the requirements of subsection 141.01(5) of the Act and that the appellant can resort to the computation authorized by subsection 21(2) of the Regulations for the 1991, 1993 and 1994 periods. In light of this result, I would award costs in favour of the appellant here and below.

             "Marc Noël"

                                                          J.A.

"I agree

Robert Décary J.A."

"I agree

Gilles Létourneau J.A."

Certified true translation

Suzanne M. Gauthier, trad. a., LL.L.

FEDERAL COURT OF CANADA

APPEAL DIVISION

NAMES OF COUNSEL AND SOLICITORS OF RECORD

FILE NO:                                  A-829-99

CORAM:                                  DÉCARY

LÉTOURNEAU

NOËL, JJ.A.

STYLE:                                      VILLE DE MAGOG

Appellant

and

HER MAJESTY THE QUEEN

Respondent

PLACE OF HEARING:         Montréal, Quebec

DATE OF HEARING:            May 16, 2001

REASONS FOR JUDGMENT OF THE COURT BY NOËL J.A.

DATED:                                    June 21, 2001

APPEARANCES:

Serge Fournier                                                                  FOR THE APPELLANT

Michel Morel                                                                      FOR THE RESPONDENT

SOLICITORS OF RECORD:

Brouillette Charpentier Fortin

Montréal, Quebec                                                             FOR THE APPELLANT

Veillette & Associés Contentieux

Montréal, Quebec                                                             FOR THE RESPONDENT