Bowman T.CJ.:
These appeals are from assessments for five taxation years of the appellant ending December 31, 1985, December 31, 1986, June 30, 1987, December 31, 1987 and December 31, 1988. In respect of two of the three issues raised in the notice of appeal the appellant has conceded the correctness of the respondent’s position, viz., the issue relating to the appellant’s claim to deduct in computing taxable income for 1985 and subsequent years a non-capital loss incurred in 1982 and the issue relating to the Minister’s refusal to include an amount of $1,118,730 in the appellant’s opening balance for 1985 and subsequent years the undepreciated capital cost of certain property. It is agreed that there will be no costs with respect to these issues.
The sole remaining issue is the treatment of expenditures incurred in the years in question by the appellant and claimed by it as expenses of qualifying scientific research and experimental development (“SR & ED”).
The appellant is a large company engaged in the production and sale of woven fabrics. That business may, for the purposes of these appeals, be divided in four parts: fashion, outerwear, industrial and print. That it carries on SR & ED as a necessary part of its business is undisputed. The issue is the computation of the SR & ED expenditures.
The amounts involved are set out in the notice of appeal, as corrected by the appellant at trial:
TAXATION | | CLAIMED | ALLOWED |
YEAR | |
ENDING | |
December 31, | SR & ED | $ 2,453,723 | $ 1,089,453 |
1985 | |
| ITC | 490,745 | 217,891 |
December 31, | SR & ED | 2,833,410 | 1,258,034 |
1986 | |
| ITC | 557,383 | 251,607 |
June 30, 1987 | SR & ED | 1,417,509 | 629,374 |
| ITC | 278,168 | 125,875 |
December 31, | SR & ED | 1,487,786 | 660,577 |
1987 | |
| ITC | 285,119 | 132,115 |
December 31, | SR & ED | 2,485,829 | 1,103,708 |
1988 | |
| ITC | 497,166 | 220,742 |
It is not disputed that the amounts were spent. What is disputed is that they all qualify as SR & ED expenditures. It is also common ground that they are of a current rather than a capital nature. The effect of the expenditures being, or not being, SR & ED expenditures is as follows:
(a) if they are not they are nonetheless either currently deductible under section 9 of the Income Tax Act or form part of the cost of inventory;
(b) if they are current SR & ED expenditures:
(i) they may be deducted in the year of expenditure under section 37 or they may, at the taxpayer’s option, be put into an SR & ED pool and deducted over a period of time; and, in any event,
(ii) they qualify for an investment tax credit (“ITC”), which, in the years in question, was 20%.
The statutory basis for the claim that SR & ED expenses are deductible is as follows:
Subsection 37(1):
(1) Where a taxpayer carried on a business in Canada in a taxation year and files with his return of income under this Part for the year a pre- scribed form containing prescribed information, there may be deducted in computing his income from the business for the year such amount as he may claim not exceeding the amount, if any, by which the aggregate of
(a) the aggregate of all amounts each of which is an expenditure of a current nature made by the taxpayer in the year or in a preceding taxation year ending after 1973
(i) on scientific research and experimental development carried on in Canada, directly undertaken by or on behalf of the taxpayer, and related to a business of the taxpayer, or
exceeds the aggregate of
(d) the aggregate of all amounts each of which is the amount of any government assistance or non-government assistance (within the meanings assigned to those expressions by subsection 127(9)) in respect of an expenditure described in paragraph (a) or (b) that, at the time of filing of the return of income for the year, the taxpayer has received, is entitled to receive or can reasonably be expected to receive;
(e) that part of the aggregate of all amounts each of which is an amount deducted under subsection 127(5) in computing the tax otherwise payable by the taxpayer under this Part for a preceding taxation year that may reasonably be attributed to expenditures of a current nature made in a preceding taxation year that were qualified expenditures in respect of scientific research and experimental development for the purposes of section 127;
(f) the aggregate of all amounts each of which is an amount deducted under this subsection in computing the taxpayer’s income for a preceding taxation year, except amounts described in subsection (6);
(g) the aggregate of all amounts each of which is an amount equal to twice the amount claimed under subparagraph 194(2)( a)(ii) by the taxpayer for the year or any preceding taxation year, and
(h) where the taxpayer is a corporation control of which has been acquired by a person or group of persons before the end of the year, the amount determined for the year under subsection (6.1) with respect to the corporation.
(Paragraphs (d) to (h) have no application but counsel referred to them in the course of argument.)
Paragraph 37(7)(b) reads as follows:
(b) “scientific research and experimental development” has the meaning given to that expression by regulation.
Paragraph 37(7)(c) reads as follows:
(c) references to expenditures on or in respect of scientific research
(i) where the references occur in subsection (2), include only
(A) expenditures each of which was an expenditure incurred for and all or substantially all of which was attributable to the prosecution of scientific research and experimental development, and
(B) expenditures of a current nature that were directly attributable, as determined by regulation, to the prosecution of scientific research and experimental development, and
(This subparagraph refers to subsection 37(2), which deals with research outside Canada and is therefore inapplicable.)
(11) where the references occur other than in subsection (2), include only
(A) expenditures each of which was an expenditure incurred for and all or substantially all of which was attributable to the prosecution, or to the provision of premises, facilities or equipment for the prosecution, of scientific research and experimental development in Canada, and
(B) expenditures of a current nature that were directly attributable, as determined by regulation, to the prosecution, or to the provision of premises, facilities or equipment for the prosecution, of scientific research and experimental development in Canada.
Subparagraph (i) is not relevant because the claim is made under subsection 37(1). Clauses (A) and (B) of subparagraph 37(7)(c)(ii) are relied upon as applicable to the appellant’s claim. Clause (A) could cover both capital and current expenditures.
For the purposes of claiming the ITC under subsection 127(5), “qualified expenditure” is defined in subsection 127(9) as follows:
“qualified expenditure” means an expenditure in respect of scientific research and experimental development made by a taxpayer after March 31, 1977 that qualifies as an expenditure described in paragraph 37(1)(«) or subparagraph 37(l)(h)(i), but does not include
(a) a prescribed expenditure, nor [sic]
(b) in the case of a taxpayer that is a corporation, an expenditure specified by the taxpayer for the purposes of clause 194(2)(«)(ii)(A).
SR & ED is defined in subsections 2900(1) and (2) of the Income Tax Regulations as follows:
(1) For the purposes of this Part and paragraphs 37(7)(6) and 37.1(5)(e) of the Act, “scientific research and experimental development” means systematic investigation or search carried out in a field of science or technology by means of experiment or analysis, that is to say,
(a) basic research, namely, work undertaken for the advancement of scientific knowledge without a specific practical application in view,
(b) applied research, namely, work undertaken for the advancement of scientific knowledge with a specific practical application in view, or
(c) development, namely, use of the results of basic or applied research for the purpose of creating new, or improving existing, materials, devices, products or processes,
but does not include activities with respect to
(d) market research or sales promotion;
(e) quality control or routine testing of materials, devices or products;
(f) research in the social sciences or the humanities;
(g) prospecting, exploring or drilling for or producing minerals, petroleum or natural gas;
(h) the commercial production of a new or improved material, device or product or the commercial use of a new or improved process;
(i) style changes; or
(j) routine data collection.
(2) For the purposes of clauses 37(7)(c)(7)(B) and (h)( ) of the Act, the following expenditures are directly attributable to the prosecution of scientific research and experimental development:
(a) the cost of materials consumed in such prosecution;
(b) where an employee directly undertakes, supervises or supports such prosecution, the portion of the salaries or wages and related benefits paid to or for that employee that can reasonably be considered to relate thereto; and
(c) other expenditures that are directly related to such prosecution and that would not have been incurred if such prosecution had not occurred.
To put the problem in a nutshell, the appellant in 1988 and prior years carried on SR & ED on a substantial scale in connection with the textiles business in which it was engaged. It sold some of the products that were produced in the course of that development -- evidently in some cases for proceeds that exceeded the cost of the SR & ED expenditures. Different approaches were considered by the Department of National Revenue, specifically:
(a) should the sale proceeds of the fabrics produced in the course of prosecuting the SR & ED be applied against the cost of SR & ED (the “netting” approach)?
(b) should some portion of the costs claimed as SR & ED expenditures be attributed not to SR & ED but to the cost of inventory (the “carve out” approach)? One of the premises on which this approach is based is that the appellant in the course of its development produced more material than was needed for its development purposes and accordingly some portion of the cost should be attributed to production.
The parties agreed that the case would proceed on the basis that only the appeal for 1988 would be argued, and the result of the appeal for that year would apply to the other years. Moreover, for that year, ten projects were chosen by the parties — five by the appellant and five by the respondent
—— and it was agreed that they would be taken as representative.
Preliminary question — The agreement of January 15, 1992. (L4/R646/T0/BT0) 1.011 1248_4827_4961
On January 15, 1992, the appellant’s Vice-President, Finance, Mr. Barry Yager, signed a document in which he agreed to a formula for determining the allowable portion of the claimed SR & ED costs. The Minister assessed the years in question on that basis. The preliminary question is whether that agreement is a bar to the appellant’s pursuing these appeals. The audit for the taxation years 1985 to 1988 had been proceeding for a considerable period of time. Representations had been made to Mr. O’Grady, the assessor in the Montreal District Office of the Department of National Revenue. On July 10, 1991 Price Waterhouse, the appellant’s auditors, wrote a lengthy submission to Mr. O’Grady. There ensued an exchange of memoranda between him and the head office of the Department of National Revenue but no final resolution of the matter was achieved. On January 15, 1992 the auditor for the appellant, Mr. Di Palma, and the Vice-President of Finance of the appellant, Mr. Barry Yager, met with Mr. O’Grady.
At that time they presented him with a document setting out the SR & ED expenditures of the appellant. The document is reproduced as Schedule I.
The meeting took place in Mr. Yager’s office and it lasted about an hour. Mr. Di Palma was aware of the exchange of memoranda between Mr. O’Grady and the head office of the Department in Ottawa. The appellant was concerned that the Department might assess using the netting method, by which the proceeds of sales of fabric produced in the experimental stage would be set off against SR & ED expenditures, reducing them to nil or, alternatively that they might “carve out” 65% of the SR & ED expenditures as part of the cost of sales.
Mr. Di Palma’s testimony with respect to the circumstances surrounding the agreement was as follows:
Q. And what was the essence of the meeting, what transpired?
A. What transpired was that we were six or seven months after a request was made to Ottawa, we were three or four months after a response was received from Ottawa, and we were almost a year from the time when Revenue Canada finished their field work, and yet, Revenue Canada was not in a position to reassess us because they had no numbers to reassess us on.
They knew that we were preparing this carve-out sheet as it was also sent to Revenue Canada, to my knowledge, on, in December, and the purpose of the meeting was, where do we go from here? It was told to us that Revenue Canada was not going to allow us a hundred percent (100%) of the R & D expenditures.
Q. That was clear to you?
A. That was clear, that was, it was a non-issue. They have, but the fact of the matter is we were almost two years after the audit commenced, they had to reassess on some basis. So, these numbers were given to them because it was the best scenario for them to reassess which cost out-of-pocket the taxpayer the least amount of money, because I was worried and I advised the taxpayer that they could reassess using their so-called sixty-five percent (65%), which he makes reference to in his letter from Ottawa, and the fact that they made reference to the I.T. about netting the sales.
Q. The entire sales which would wipe out?
A. It would wipe out most of our eighty percent (80%) at least of our R & D.
Q. Right. Now did, at the meeting, in the course of the meeting, was an agreement reached?
A. Agreement was reached for Revenue Canada to allow them to reassess on a certain basis, because they did not come to us on a potential reassessment basis, we gave it to them, it was for purposes of them reassessing us.
Q. And the basis was as described in A-l, in Tab --
A. Tab 1, yes of A-2.
(This refers to the document reproduced as Schedule 1 to these reasons.)
Q. Tab 1 of A-2, I’m sorry, and during the course of the meeting, was Mr. Yager requested to sign a document?
A. Mr. O’Grady said that in order for them to reassess, they need some permission from the taxpayer to reassess on a certain basis. I presume, I don’t know, but I presume because there was nothing other than this one sheet supporting a percentage that he needed a letter from the taxpayer to allow them to reassess them on a certain basis.
A. The copy in my files is the equivalent to the one that is put into the Court, except for, I do not have this forty-four point four percent (44.4%) adjustment, but I know what it relates to and it’s proper in its context of this letter, and I do not have the comments at the bottom confirming Barry Yager, January fifteenth
(15th) at three O nine P.M. (15 H 09).
Q. Right, and does your copy though in your file contain other indication that is not in this document, in this report?
A. Yes, once copy at the meeting, copies of this letter were distributed, I took some notes down, I put down “to attend at the meeting”, obviously Barry Yager asked me: “What is the problem with me signing this?” And I was at the point, as an advisor, which I was, I had never been before in my life, is that we were giving Revenue Canada basis to reassess because they did not come to us on a basis to reassess. However, we had no agreement anywhere and they needed to reassess, that was obvious, and they were going to reassess.
So, I said, this is the worst of all evils for them to reassess and we have nothing to lose by signing this, if something comes up, we can always object, we have ninety (90) days from the date that we have, that we get a notice of reassessment to object.
Q. And that is the advice you gave to Mr. Yager?
A. That is the advice I gave to Mr. Yager.
Q. Did you give that advice in front of Mr. O’Grady?
A. I told him we had, in front of Mr. O’Grady, that we have nothing to lose by signing this because we can always object if something comes up.
Q. And did Mr. O’Grady say, well --.
A. He didn’t say anything.
Exhibits A-1 and A-7 of the agreement were written and signed by Mr. Yager. Exhibit A-4 was the copy kept by Mr. O’Grady. It reads as follows:
Revenue Canada: (L504/R3950/T2/BT2) test_marked_paragraph_end (3942) 0.831 1250_7491_7595
I agree to the 56% basis for establishing allowable R&D expenditures in taxation years 1985 to 1988.
B. Yager
Vice-President Finance
Montreal
January 15, 1992
On this copy 56% is struck out and 55.6% is written and initialed by Mr. Yager. That figure of 55.6% is struck out and “*44.4%” is substituted in Mr. O’Grady’s handwriting and at the bottom, again in Mr. O’Grady’s handwriting, is written:
confirmed with Barry Yeager [sic] January 15, 1992 at 3:09 p.m.
D.O.G.
Also, beside Mr. O’Grady’s signature are the initials J.R. but Mr. O’Grady was unsure whose initials they were.
Exhibit A-7 came from Mr. Di Palma’s files. The text of the agreement is, of course, the same as Exhibit A-4. The substitution of 44.4% on Mr. O’Grady’s copy (Exhibit A-4) is not present nor, of course, is Mr. O’Grady’s comment about confirming the change with Mr. Yager. On Mr. Di Palma’s copy, in addition to the names of those present at the meeting (Mr. Di Palma (PW), Antoinelle Lapolla and Barry Yager (CI) and Dennis O’Grady (RC)), are the following notations by Mr. Di Palma:
-I advised the taxpayer that we could appeal in any event and Revenue Canada did not disagree.
-They agreed it was R&D but sales problem.
I find the notation that he advised the taxpayer that it could appeal selfserving and I expressly refrain from making any finding that Mr. Di Palma told Mr. Yager that the appellant could appeal, that he did so in Mr. O’Grady’s presence or that Mr. O’Grady heard him say so, if it was said at all. Mr. Yager was not called. In the final analysis, however, what Mr. Di Palma may have told Mr. Yager about his legal rights of objection after signing the agreement is not germane to the question here.
In the result, Mr. O’Grady accepted the offer of Consoltex and an agreement was formed. The resulting assessment was based upon the agreement and upon the figures contained in Schedule I.
That document requires some explanation. The notation at the bottom of it “settled at 55.6%” was evidently made by someone at Price Waterhouse or by someone in the appellant’s employment because the copy of the document (Exhibit A-2) came from the appellant’s records. The manner in which the 55.6% figure was arrived at was as follows:
One starts with the SR & ED expenses claimed by the appellant. From that, one deducts the cost of sales of the products sold. In the case of fashion goods this is $821,969 minus $531,074 leaving a “Net R&D expense” of $290,895.
The cost of sales of SR & ED products ($531,074) is arrived at as follows:
The standard cost of producing an equivalent amount of fabric in the production phase is determined ($772,809). From that is deducted depreciation ($43,277), cost of seconds ($54,097 and $11,592) to arrive at a net cost of sales of the goods produced in the SR & ED stage of $663,843. Both parties accepted an assumption that 80% of the SR & ED production was sold, and therefore 80% of $663,843 was determined to arrive at $531,074. For the purposes of the settlement, the total amount claimed as SR & ED expenditures, $821,969, was reduced by $531,074 being the amount assumed to be the portion attributable to cost of sales, leaving $290,895 as SR & ED expenditures in respect of fashion fabrics.
The same calculation was performed in respect of outerwear and print fabrics to arrive at corresponding figures of $385,729 and $76,176 respectively and to those were added direct expenses (for example salaries of persons engaged exclusively in SR & ED) to arrive at a total of $1,103,788. To this was applied the 20% investment tax credit rate to arrive at $220,758. This works out to 44.6% of the amount claimed by the appellant.
These calculations were apparently designed to find an acceptable basis of determining a formula premised upon the appropriateness of the “carve out” method. Mr. Di Palma agreed in his testimony that if the principle of carve out is appropriate the methodology used in arriving at the 56.6%/44.4% split between SR & ED and cost of sales is reasonable. If I believed that as a matter of law the principle of carve out were appropriate I daresay this formula would be as good as any other.
On the basis of this agreement the Minister ceased any further consideration of the appellant’s claim for ITCs on its SR & ED expenditures and assessed accordingly. I find as a fact that the Minister accepted and relied upon the settlement proposed by the appellant and based the assessments for 1985, 1986, 1987 and 1988 upon it. Mr. Yager offered the settlement, Mr. O’Grady accepted it and there was thereby a completed agreement under which the Minister fulfilled his part of the bargain by issuing the assessments in question. I believe that Mr. O’Grady acted responsibly and in good faith in accepting the offer and that he saw it as a complete resolution of the matter and not as an interim step which would merely accelerate an assessment to which objection could be taken. Since Mr. Yager did not testify it is impossible to know whether he saw the agreement as merely one to assess or one that was intended as a final resolution of the matter. In light of the conclusion that I have reached on the basis of Cohen (infra) it is irrelevant how the parties saw it because in either case they were not bound by it.
The reassessments were issued on May 1, 1992 and, following objection to and confirmation of the reassessments, the matter came before this court.
The objections were filed after the decision of Judge Lamarre Proulx in the case of Cultures Laflamme (1984) Inc. c. Ministre du Revenu national, (1992), 93 D.T.C. 603 (T.C.C.) although Mr. Lefebvre informed the court, as counsel, that he had been consulted prior to that decision about objecting. Nothing turns on the timing of the objection. The question is whether, as a matter of law, the appellant is bound by the agreement. The decision of the Federal Court of Appeal in Cohen v. R., (1980), 80 D.T.C. 6250 (Fed. C.A.) is clear authority for the proposition that where the Minister and the taxpayer make an agreement as to the manner in which the taxpayer is to be assessed, the Minister is not bound by the agreement and may renege on the deal if he chooses to, even though the other party has acted to his or her detriment on the basis of the agreement.
In Cohen, Mr. Nathan Cohen and his partner Mr. Zalkind agreed with the officials of the department that they would accept that the sale of certain property was a transaction on revenue account and that the sale of other lands (“the Bourret lands”) was on capital account. In reliance upon the agreement the taxpayer did not object to the treatment of the proceeds from the sale of the lands which he had agreed would give rise to income. The Minister then, contrary to the agreement and after the time for objecting had expired, assessed the gain on the Bourret lands as income. The Honourable Senator Lazarus Phillips, Q.C. testified at trial that there was such an agreement and there is no suggestion in the reasons for judgment of the trial judge, Décary J., ((1978), 78 D.T.C. 6099 (Fed. T.D.)) that he did not accept that there was such an agreement. He held however that the Minister could ignore the agreement.
In the Federal Court of Appeal Pratte J. said:
The appellant’s second argument was that the Minister could not legally reassess the appellant on the basis that the profit in question was income because he had previously agreed to treat that profit as a capital gain. Counsel submitted that his agreement had been made during the course of negotiations between representatives of the appellant and officers of the Department of National Revenue concerning the appellant’s assessments for the years 1961 to 1964. The appellant had agreed, said counsel, not to appeal his assessments for the 1961 to 1964 taxation years on the understanding that his income tax for 1965 would be computed on the basis that the profit here in question was a capital gain. Counsel argued that the Minister could not repudiate that understanding, particularly after the expiry of the time within which the appellant might have appealed the 1961 to 1964 assessments.
In my view, the Trial judge correctly dismissed that argument, “...that Minister has a statutory duty to assess the amount of tax payable on the facts as he finds them in accordance with the law as he understands it. It follows that he cannot assess for some amount designed to implement a compromise settlement...” (Galway v. Minister of National Revenue [1974] 1 F.C. 600 at page 602, [74 D.T.C. 6355 at page 6357]). The agreement whereby the Minister would agree to assess income tax otherwise than in accordance with the law would, in my view, be an illegal agreement. Therefore, even if the record supported the appellant’s contention that the Minister agreed to treat the profit here in question as a capital gain, that agreement would not bind the Minister and would not prevent him from assessing the tax payable by the appellant in accordance with the requirements of the statute.
Counsel at trial had argued that the Minister was “estopped” from reassessing as he did. I do not think that it was an appropriate use of the term. The word “estoppel” is frequently used in situations to which it does not apply. In Goldstein v. R, (1995), 96 D.T.C. 1029 (T.C.C.), at 103-1034 the matter was discussed as follows:
There is much authority relating to the question of estoppel in tax matters and no useful purpose would be served by yet another review of the cases. I shall endeavour however to set out the principles as I understand them, at least to the extent that they are relevant. Estoppels come in various forms -- estoppel in pais, estoppel by record and estoppel by deed. In some cases reference is made to a concept of “equitable estoppel”, a phrase which may or may not be accurate . It is sufficient to say that the only type of estoppel with which we are concerned here is estoppel in pais. In Canadian Superior Oil Ltd. v. Paddon- Hughes Development Co. Ltd. [1970] S.C.R. 932 at 939-940 Martland J. set out the factors giving rise to an estoppel as follows:
The essential factors giving rise to an estoppel are I think:
(1) A representation or conduct amounting to a representation intended to induce a course of conduct on the part of the person to whom the representation is made.
(2) An act or omission resulting from the representation, whether actual or by conduct, by the person to whom the representation is made.
(3) Detriment to such person as a consequence of the act or omission.
Estoppel is no longer merely a rule of evidence. It is a rule of substantive law. Lord Denning calls it “a principle of justice and of equity”.
It is sometimes said that estoppel does not lie against the Crown. The statement is not accurate and seems to stem from a misapplication of the term estoppel. The principle of estoppel binds the Crown, as do other principles of law. Estoppel in pais, as it applies to the Crown, involves representations of fact made by officials of the Crown and relied and acted on by the subject to his or her detriment. The doctrine has no application where a particular interpretation of a statute has been communicated to a subject by an official of the government, relied upon by that subject to his or her detriment and then withdrawn or changed by the government. In such a case a taxpayer sometimes seeks to invoke the doctrine of estoppel. It is inappropriate to do so not because such representations give rise to an estoppel that does not bind the Crown, but rather, because no estoppel can arise where such representations are not in accordance with the law. Although estoppel is now a principle of substantive law it had its origins in the law of evidence and as such relates to representations of fact. It has no role to play where questions of interpretation of the law are involved, because estoppels cannot override the law.
I do not think that the question of estoppel had anything to do with the Cohen case, nor does it have any application here.
The result of the decision in Cohen is that the Minister is free to repudiate any agreement that he has made with respect to the manner in which he assesses a taxpayer. It follows necessarily that a taxpayer who has made a deal with the Minister is equally free to do so. Any conclusion that the Min- ister is not bound by agreements but the taxpayer is would be wholly unacceptable as a matter of principle. Binding agreements must be premised upon mutuality and reciprocity of obligations between the parties. There can obviously be no agreement where one party is bound and one is not.
In Cohen it was stated that the agreement “whereby the Minister would agree to assess income tax other than in accordance with the law would be an illegal agreement” and that... “that agreement would not bind the Minister”. As noted above, neither would the taxpayer be bound by an “illegal agreement”. If “illegal agreement” means an agreement that results in an assessment that is not precisely the same as that which would be sanctioned by a court if the matter were litigated it follows that no compromise or settlement of income tax disputes can ever be made. The basis upon which most income tax disputes are settled is that there may be some doubt concerning the result. If the only “legal” agreement is one that a court would ultimately sanction there would be no way in which the binding effect of such an agreement could be determined without litigation and this would defeat the purpose of settling, and the courts would be flooded. The vast majority of income tax disputes are settled, whether at the assessment level, on objection or after an appeal has been instituted. The system simply would break down if all proposed settlements had to be litigated. The Cohen case is a good example of the type of settlement that is routinely made, where several transactions are involved in what are commonly called “trading cases” and the parties agree that some will be treated as being on revenue account and some on capital account. According to Cohen such settlements are illegal.
Smerchanski c. Ministre du Revenu national, (1977), 76 D.T.C. 6247 (S.C.C.) was a decision of the Supreme Court of Canada. There, the taxpayer, in order to avoid a threatened prosecution for evasion of income tax, agreed in writing to pay such sums as the Minister should assess, without demanding particulars, and waiving his right to appeal. After the Minister assessed, the taxpayer paid the amounts and shortly after the time for instituting criminal proceedings expired, the taxpayer appealed against the assessments. The court held that the taxpayer, having waived his right of appeal, and, despite the underlying threat of prosecution, in the absence of bad faith, malice, or undue severity on the part of the tax authority, was not entitled to prosecute his appeal.
The Crown argued, inter alia, estoppel but the court did not consider it necessary to deal with the point. The court based its decision essentially upon the fact that the taxpayer had waived his right of appeal, and upon the taxpayer’s failure to make out the factual underpinning of his case that there had been duress and undue influence. Moreover there was some point made at trial that the taxpayer had altered documents that the Crown had returned to him. I presume that it is this to which the Supreme Court was referring when it said that even assuming the waivers were voidable Mr. Smerchan- ski’s conduct would disentitle him to any relief.
None of these factors exist here. There was no express waiver of a right of objection or appeal in this case, no threat of a criminal prosecution and no allegation of undue influence. I must assume that the Federal Court of Appeal in Cohen was aware of the Smerchanski judgment. I cannot therefore ignore Cohen on the basis that it is inconsistent with Smerchanski?
Professors Hogg and Magee in Principles of Canadian Income Tax Law (Toronto: Carswell 1995) at page 21 state:
The effect of the Smerchanski and Cohen cases is that the taxpayer is bound by a settlement agreement, but the Minister is not. Of course, a settlement of litigation that was implemented by a formal entry of judgment would then have the force of a court judgment, which is binding on both parties. However, in Galway v. Minister of National Revenue (1974), the Federal Court of Appeal refused an application for a consent judgment to implement the terms of a settlement agreement between the Minister and a taxpayer. According to the Court, the Minister has no power to assess in accordance with a “compromise settlement”, and the Court should not sanctify an ultra vires act. The Minister’s duty is to assess in accordance with the law, and the only kind of settlement that the Court would be prepared to implement by a consent judgment would be one in which the parties were agreed on the application of the law to the facts.
The attitude of the Federal Court of Appeal in Cohen and Galway is far too rigid and doctrinaire. If the Minister were really unable to make compromise settlements, he or she would be denied an essential tool of enforcement. The Minister must husband the Department’s limited resources, and it is not realistic to require the Minister to insist on every last legal point, and to litigate every dispute to the bitter end. Most disputes about tax are simply disputes about money which are inherently capable of resolution by compromise. Presumably, the Minister would agree to a compromise settlement only on the basis that it of- fered a better net recovery than would probably be achieved by continuance of the litigation. It seems foolish to require the Minister to incur the unnecessary costs of avoidable litigation in the name of an abstract statutory duty to apply the law.
In general, I agree with their observations, subject to one qualification. I do not think that the Smerchanski and Cohen cases, read together, can be taken to justify a conclusion that the taxpayer is bound by a settlement agreement but the Minister is not. It is unconscionable enough that the Minister should be able to renege on settlements that he or she has made. It would be doubly indefensible that a taxpayer should be unilaterally bound to honour agreements that the Minister is free to repudiate. Neither the Minister nor the appellant is bound by the agreement of January 15, 1992. Of course the Minister acted on the agreement by assessing in accordance with it, but this does not distinguish the case from Cohen, because Mr. Cohen as well implemented the agreement by refraining from objecting to the first assessment. There are three possible alternative and inconsistent results of the Cohen, Galway and Smerchanski decisions: (a) the taxpayer and the Minister are both bound by such agreements; (b) neither is bound; and (c) the taxpayer is bound but the Minister is not. Assuming that Cohen is correct in law, so that (a) cannot apply, the least unacceptable result of the two remaining alternatives is (b).
One may approach the matter from a slightly different perspective and arrive at the same result. Assuming Cohen to be correct in law, it follows that the Minister lacks the capacity to make such settlements. If one party to a deal lacks the capacity to do so, there is no deal.
I have therefore reluctantly concluded that the appellant is not bound by the agreement signed on January 15, 1992 by its Vice-President of Finance.
I come now to a consideration of the case on its merits, unfettered by any question of an agreement.
As stated, the appellant carries on extensive scientific research in developing new types of woven fabrics. The problem here, and the reason the matter is before the court, is attributable to two facts:
(a) The research is done in the production facilities of the appellant. There is no separate research facility. This raises no discrete question of principle — research is research wherever it is carried on — but it does render it more difficult to draw a clear line of demarcation between production and research.
(b) Some — probably the larger part — of the fabrics produced in the course of the research and development are sold. Indeed, the proceeds from such sales exceed in total the SR & ED expenses claimed.
The amounts claimed as SR & ED were in fact spent. This is not disputed. If the work done was all SR & ED within the meaning of section 37 of the Income Tax Act and Part XXIX of the Regulations had there been no sales of the textiles produced in the course of the scientific research, does it become any the less SR & ED because some of the product was sold? I see no logic in such a position. Whether experimental scientific research meets the criteria of the Act and Regulations is a matter to be objectively determined irrespective of what is done with the product, whether it be sold, stored or scrapped. The disposition of the product does not impinge on the nature of the activity whereby it is created.
The more relevant question is whether, assuming the activities are all SR & ED, all or any portion of their cost is to be reduced by the proceeds of sales of product resulting from the SR & ED. Before I come to that question I must first determine whether the activities all qualify as SR & ED.
The relevant portions of subsection 2900(1) of the Regulations read:
For the purposes of this Part and paragraphs 37(7)(b) and 37.1(5)(e) of the Act, “scientific research and experimental development” means systematic investigation or search carried out in a field of science or technology by means of experiment or analysis, that is to say,
(a) ...[not relied upon by the appellant]
(b) ...[not relied upon by the appellant]
(c) development, namely use of the results of basic or applied research for the purpose of creating new, or improving existing materials, devices, products or processes.
The parties agreed that ten projects undertaken in 1988 would be treated as typical of all projects undertaken in all the years under appeal. I do not propose to describe these projects in the detail in which they were described in the viva voce and documentary evidence. It was not seriously disputed that all of the projects chosen involved SR & ED. Rather, the respondent’s principal focus was on the amount of fabric produced, the effect of the sales and the effect of subparagraph 37(2)(c)(ii) and section 2900 of the Regulations. I shall, however, briefly describe the projects.
The ten projects are set out in Exhibit A-1 and are preceded by an expert witness report by Dr. Arthur D. Broadbent. In that report he outlined the basic textile operations of the appellant. The appellant does not manufacture yarn. It acquires it from outside sources. Essentially the textile processing involves two stages: weaving and converting or finishing. The first stage comprises four operations:
(i) beaming or warping, which is the preparation of a set of parallel yarns wound onto a cylindrical beam;
(ii) slashing or sizing, which involves coating each yarn with a film of size to protect it against abrasion and breaking during the weaving process. The warp yarns are wound off the beam and passed through a bath of chemicals and then dried. They are then rewound onto a beam. A wide variety of polymers, chemicals and other additives are applied to them. Before the converting operation the size must be removed;
(iii) entering, which involves threading the yarns through a series of guides mounted in frames called heddles which are placed on the loom and raised or lowered in a predetermined sequence to create a gap between the upper and lower layer of warps through which the filling yarn will be inserted;
(iv) weaving, which is a complex operation, but at the risk of oversimplification, it may be described as the insertion of filling yarns between the warp yarns, using a variety of techniques, including a traditional shuttle, a projectile or rapier that grips the yarn and water or air jets.
The second stage is converting, or dyeing and finishing. This comprises three stages:
(1) preparation, or the removal from the fabric of chemicals applied to the yarns;
(ii) dyeing, a complex chemical process involving the subjecting of the fabric by a variety of techniques, to chemical dyes; and
(iii) finishing, which is the mechanical, chemical or thermal treatment of the dyed fabric to give it particular physical characteristics.
The first five projects set out below were chosen by the appellant. They fall into three categories — fashion, outerwear and industrial.
Development 6024 (Fashion) (L16/R3344/T0/BT0) test_marked_paragraph_end (2474) 1.011 1260_8447_8599
The purpose of this project was to develop summer-weight fabrics with a new %o polyester/viscose slub yarn. It was one of the first projects involving spun yarns rather than continuous filaments. It was a weak yarn. 5870 metres were woven and 800 metres were dyed. There were problems with both the recovering and the dyeing. The project failed and was abandoned.
Development 5204 (Outerwear) (L12/R3118/T0/BT0) test_linespace (274>257.84) 1.034 1261_1613_1741
The purpose of this project was to improve the crinkled appearance of an existing product by using a new nylon filling yarn. No significant improvement was achieved and the project was abandoned. About 3,600 metres of greige fabric was woven.
Development 5105 (Outerwear) (L16/R3116/T0/BT0) test_linespace (274>258.02) 1.035 1261_2963_3091
The object of this development was to produce a new range of super strong fabric using very fine nylon yarn with a high thread density to give water resistance to the fabric. About 16,000 metres of greige fabric were produced. An alternative sizing technique was used to permit the use of water jets rather than air jet looms. Ultimately the fabric went into production.
Development 2169 (Industrial) (L16/R3188/T0/BT0) test_linespace (274>256.00) 1.035 1261_4747_4875
The purpose here was to develop a fire resistant nomex fabric to be used by oil drilling workers, with weight and heat resistance meeting rigid specifications. About 6,000 metres were woven using one kilometer of yarn and about 1,500 metres using another. There were problems with dyeing and also with the original warp and filling yarns. The fabric went into production.
Development 2146 (Industrial) (L14/R3188/T0/BT0) test_linespace (274>257.10) 1.035 1261_6535_6663
The object of this development was to produce a fabric of reduced weight to satisfy ballistic requirements for bullet proof jackets. The appellant was unable to reduce the weight, although, using 20 layers, the fabric was effective in stopping bullets. Therefore when it went into production it was sold as protection for hockey players.
Projects selected by the Department of National Revenue (L10/R1022/T0/BT0) test_linespace (420>261.82) 1.023 1261_8093_8221
Development 5931 (Fashion)
The purpose of this project was to improve the lustre of an existing product by replacing the texturized polyester filling yarn with a more expensive non-textured yarn. A new warp construction was required and five different dobby patterns were developed. There were numerous problems with the weaving and 20,000 metres were woven and 1,340 were finished. The development went into production.
Development 6123 (Fashion)
This involved an attempt to develop a new blouse fabric using a very fine Japanese filling yarn. Numerous technical problems were encountered with the weaving, but none with the dyeing and finishing and several machine trials were needed to establish correct weaving procedure. Simultaneous tests were done on four looms. Twenty thousand metres were woven but only 404 were finished. The development went into production.
Development 6042 (Fashion)
The object here was to reproduce existing patterns with a new type of yarn and a new warp construction at a lower yarn density. Numerous problems were encountered and about 15,000 metres were produced. Several hundred meters of fabric had to be woven between each loom adjustment. The development went into production.
Development 2182 (Industrial)
The purpose of this development was to reduce the weight of a fabric used in the outer shell of fire fighters’ suits. The new yarn required special sizing procedures and special loom adjustments. The fabric also needed to be breatheable and its weight had to be reduced. Tests had to be performed with various types of coating. Initially the project failed to achieve the required objectives of weight and permeability but ultimately it succeeded. About 3,000 metres were woven.
Development 5110 (Outerwear)
This involved an attempt to use a nylon warp yarn as the filling yarn in production of a lining fabric. The idea was to determine whether yarns with broken filaments could be used for warp construction and as wefts. Initially 2,000 metres of fabric were requested for research and development purposes, but so many problems were encountered that ultimately 52,000 metres were produced, much of it substandard. The project was not successful and the material was ultimately sold for use as liners in sleeping bags.
I have taken these summaries from Dr. Broadbent’s report and from the report by the Department of National Revenue’s own science advisors. The two are substantially consistent and establish that each of the projects involved “systematic investigation or research carried out in a field of science or technology by means of experiment or analysis, that is to say ... development, namely use of the basic results or applied research for the purpose of creating new, or improving existing, materials, devices, products or processes”.
The expert and the science advisors appear to agree on this and my own observation of the evidence confirms that the activities described in the ten projects chosen as representative meet the criteria of scientific research. The investigation is systematic and it is carried out in a field of technology by means of experimentation using basic or applied research with a view to creating new or improving existing materials products or processes. Technology and materials in the textiles field are rapidly changing and in order to maintain a competitive position it is necessary that constant research be undertaken. In none of the projects presented to the court is there anything that is either routine or predictable. When a project is commenced it is not known whether it will succeed or what modifications need to be made to achieve the intended result.
This leaves then three problems:
(a) the question whether excessive material was woven so that the weaving went beyond what was needed for SR & ED and became part of production;
(b) the question how to treat the cost of the yarn; and
(c) what to do about the proceeds of sale of the material produced in the course of SR & ED.
(a) Excess yarn
I can understand why Mr. O’Grady was concerned about some of the lengths of fabric woven — in some cases 20,000 metres and in one 52,000. 52 kilometers is a lot of fabric. There is however no evidence that any amounts that were woven were not necessary for the experimentation that was being conducted.
The seemingly high yardages are particularly noticeable where the experimentation relates to weaving. Less material is needed where the testing has to do with conversion and finishing. I accept Dr. Broadbent’s statement in his report that large amounts of material may be used before all of the problems are identified and solved, if they are solved. Indeed, in Development 5110, 52,000 metres were produced even through originally only 2,000 were requested. The need for more became apparent only when unforeseen technical problems were encountered.
The Department’s concern about excessive yarn being used must be premised upon one or both of two hypotheses:
(i) that the researchers were so incompetent that they grossly overestimated the amount of yarn needed for an experiment, or were profligate in the use of yarn in their development work; or
(ii) the researchers were somehow in cahoots with the production people and contrived to produce more material than was needed for development work so that the cost of the excess could be treated as an SR & ED expenditure that qualified for ITC.
Neither hypothesis has any foundation in the evidence.
(b) The cost of the yarn
It is apparent form tabs 9 and 10 of exhibit A-2 that the cost of raw materials involved in SR & ED far exceeds the cost of labour and overhead, at least in the fashion and Outerwear divisions. In the fashion division for 1988, for example, raw materials cost $534,542 and labour and overhead $287,453. In the Outerwear division the proportion was $806,168 to $278,574. I would surmise that the same relative differences would exist in the other divisions, although this is not in evidence.
The treatment of the cost of raw materials necessarily used in the SR & ED is obviously a cost of the research, which could not be carried on without them. The cost of labour, considering that the research is done in the production facilities using production personnel must necessarily be allocated on some basis between production and development since the same personnel may work in development and production. These fall under clause 37(7)(c)(ii)(B) because they are “directly attributable” (as determined by regulation) to the prosecution of SR & ED and paragraph 2900(2)(b) sanctions a reasonable allocation. The cost of labour cannot fall within clause 37(7)(c)(ii)(A) because it is not an expenditure “for and all or substantially all of which was attributable to the prosecution” of SR & ED.
I have difficulty in seeing just where overhead fits into the picture at all. It does not seem to fall into clause 37(7)(c)(ii)(A), and it is questionable whether it is “directly attributable” under the meaning of clause 37(7)(c)(ii)(B) as defined in subsections 2900(2) and (3) of the Regulations. However, the point was not argued and it is not clear whether the assessment was based on an assumption that overhead is properly includible in the calculation of SR & ED expenditures. I therefore express no view on the matter.
The larger problem is the cost of yarn used in the development work. If it may be claimed under clause 37(7)(c)(ii)(A), it is a fair statement that it is an expenditure incurred for and all or substantially all of which was attributable to the prosecution of SR & ED in Canada. The fact that some of the product that resulted from the SR & ED was sold does not appear to be germane to that question. It was still a cost of the research.
If, on the other hand, the cost of yarn must, like labour, be claimed under clause 37(7)(c)(ii)(B), it is necessary that it be “directly attributable”, as determined by regulation, to the prosecution of SR & ED. To fall with the provision of subsection 2900(2) of the Regulations the cost must be either:
(a) the cost of materials consumed in such prosecution; or
(c) other expenditures that are directly related to such prosecution and that would not have been incurred if such prosecution had not been incurred.
I think it would put a strained interpretation on the word “consumed” to say that the yarn that is turned into product that is sold was “consumed”. It is precisely because it was not consumed, but was sold, that the matter is before the court. The yarn is a material that is not consumed within the meaning of paragraph 2900(2)(a). Does this exclude it from paragraph 2900(2)(c), on the theory that paragraph (a) deals with materials and if the cost of materials is to be treated as an SR & ED expenditure it must fall within paragraph (a)? Such a conclusion in my view has no basis as a matter of statutory construction. The cost of yarn is directly related to the prosecution of SR & ED and would not have been incurred if such prosecution had not occurred.
Therefore I think the cost of yarn used in the SR & ED falls within both clause 37(7)(c)(ii)(A) of the Act and paragraph 2900(2)(c) of the Regulations.
(c) The sale of the material
The Act gives special treatment to SR & ED expenditures. There is nothing that requires that such expenditures be netted against sales and the contention runs counter to the plain meaning of the word expenditure. Paragraph 37(1)(d) of the Act requires that there be deducted from SR & ED expenditures the amounts of any government assistance or non-government assistance received in respect of the expenditures. Had Parliament intended that amounts qualifying as SR & ED also be reduced by the proceeds of sales it would have been quite capable of saying so. The word “expenditures” is a reasonably comprehensible English word and “dépense” is equally straightforward in French. It is something that one spends, an outlay. No meaning of which I am aware, either in ordinary parlance or in decided cases, would justify adding after it “net of sales proceeds”. The Income Tax Act is a sophisticated régime that deals with great specificity with outlays and receipts as, for example, in the complex interaction of the resource provisions in section 59 and sections 66 and 66.8. I do not think that it is appropriate to drive a coach-and-four through the Act by requiring a netting of expenditures with sales proceeds. On this point I am in respectful agreement with the decision of my colleague Lamarre Proulx J. in Cultures Laflamme (1984) Inc. (supra).
In one sense the “carve out” approach may appear superficially to be a modified version of the netting approach, but it is based upon a fundamentally different premise. Whereas the netting approach is premised on the view that “expenditures” must be reduced by sale proceeds, the “carve out” approach is based on the theory that where the fabric produced in the course of the SR & ED operation is sold some portion of the costs of SR & ED should be attributed to cost of sales. The point has a certain facile attraction, but it does not bear close analysis. Obviously some of the cost of the yarn does form part of the cost of the goods sold. One cannot compute profit without deducting costs. That is fundamental. Mr. Marecki contended that only the marginal difference between the ordinary cost of producing fabric sold and the cost of producing such fabric in the course of SR & ED operations should be treated as an SR & ED expenditure. I do not however think that this view conforms to the scheme and purpose of the SR & ED provisions of the Act (see Highway Sawmills Ltd. v. Minister of National Revenue, (1966), 66 D.T.C. 5116 (S.C.C.), per Cartwright J. at p.5120). This is not in my view comparable to the situation with which the court was faced in Denison Mines Ltd. v. Minister of National Revenue, (1974), 74 D.T.C. 6525 (S.C.C.)where it was unsuccessfully contended that the cost of mining ore, where the ore came from what ultimately became main haulageways, was a capital cost of the haulageways and was not required to be deducted in the current cost of producing ore. Here the principal object of the expenditures is to do research and development. Moreover, it is not a question of attributing an expense to the cost of capital assets and removing it from the cost of production. The expenditures here are deductible currently as a cost of the appellant’s income producing operations. The sole question is whether they are a cost of SR & ED and as such qualify for the ITC. In my view the object of section 37 of the Act and section 2900 of the Regulations would be defeated if some portion of the SR & ED expenditures bypassed section 37 and were, in effect, put directly into cost of sales without being treated as research expenditures. Put differently, the cost of producing the fabric in the course of the research activity would, if the yarn is sold, unquestionably enter into the cost of goods sold if section 37 did not exist. Since section 37 does exist, however, they are also SR & ED expenditures and are deductible as such and are entitled to whatever beneficial treatment is accorded to them by section 37. Of course they cannot be deducted twice (subsection 4(4)). There is no basis upon which, if they qualify under the specific provisions of section 37, the fact that they otherwise would from part of the cost of inventory should require that some part or all of these expenditures should be excised from the operation of section 37. The flaw in the respondent’s reasoning lies in approaching the matter as an either/or proposition: either the expenditure is SR & ED or it is a cost of goods sold, and the two hypotheses are mutually exclusive. In fact, they are not. They are SR & ED expenditures that, as it turns out, can also be viewed as forming part of the cost of inventory. The fact that they happen to be the latter does not prevent their being the former.
In a case of this type where complex provisions of the Act are involved there is a danger that one may become entangled in technicalities and lose sight of the forest by too minute an examination of the bark on the trees. If one takes a couple of steps back and seeks to determine what the scientific research provisions of the Act are designed to accomplish, it is clear that they should be interpreted in a manner that encourages scientific research in this country. To whittle away at those provisions defeats that object. It is not, after all, as if a taxpayer who incurs current scientific research expenses is being given a double deduction or one that it would not otherwise have. The appellant would have been able to deduct these expenses even if section 37 did not exist. All it is getting is the incentive of an investment tax credit. All expenditures, whether capital or current, that give rise to ITCs are deductible one way or another in computing income, either currently or over time, and all of them, one way or another, result, or are intended to result in or the production of income. It runs counter to the entire philosophy of fiscal incentives, whether in the form of SR & ED allowances or ITCs, to say that if the expenditures giving rise to the incentives result, actually or potentially, in income they should be watered down.
The appeals are therefore allowed, with costs, and the assessments are referred back to the Minister of National Revenue for reconsideration and reassessment in accordance with these reasons.
Appeal allowed.