Coppley Noyes & Randall Ltd. v. The Queen, 91 DTC 5291, [1991] 1 CTC 541 (FCTD), varied on appeal 93 DTC 5196, 5508 (FCA).

Reed, J.:—The issue in this case is whether the plaintiff's claim that a reserve of $3,084,000 for doubtful debts for its 1982 taxation year is reasonable, as required by paragraph 20(1)(l) of the Income Tax Act, R.S.C. 1952, c. 148 (am. S.C. 1970-71-72, c. 63) (the"Act"):

20. (1) Notwithstanding paragraphs 18(1)(a), (b) and (h), in computing a taxpayer's income for a taxation year from a business or property, there may be deducted such of the following amounts as are wholly applicable to that source or such part of the following amounts as may reasonably be regarded as applicable thereto:

(I) a reasonable amount as a reserve for

(i) doubtful debts that have been included in computing the income of the taxpayer for that year or a previous year, and

(ii) doubtful debts arising from loans made in the ordinary course of business by a taxpayer part of whose ordinary business was the lending of money;

Taxpayer's Assessment

There is no doubt that the plaintiff's estimate of the amount to be attributed as a reserve for doubtful debts (or doubtful accounts) was determined in accordance with generally accepted accounting principles (GAAP"). This is clear from the evidence of Mr. McWhinnie, the vice president of Finance and the chief operating officer of Coppley Noyes and Randall, and from the evidence of Mr. Willson, the chartered accountant with the firm of Clarkson Gordon (now Ernst & Young), who was responsible for auditing the plaintiff's books. The evidence of Dr. John R. Hanna, a professor with the School of Accountancy at the University of Waterloo and previously its director, also strongly supports this conclusion.

The plaintiff is in the business of manufacturing high quality men's clothing. It also does some importing of this type of clothing and in 1982 dealt in some items of ladies’ wear as well. A substantial part of the plaintiff's business involves and involved, selling clothing to small independently owned retail shops specializing in the sale of quality men's wear. These were described as being typically" one-man shows" with the owner retailer being in the business because of his love of fine clothes. These businesses are often under capitalized, lack business experience and are very vulnerable to setbacks arising as a result of the individual characteristics of the owner retailer (e.g., lack of health) as well from negative economic conditions generally.

The plaintiff's year end is November 30. It prepares its financial statements and tax returns in the following March and April. For the purposes of both its financial statements and its tax returns for the 1982 year, Coppley Noyes reviewed its November 27, 1982 accounts receivable as of February 25, 1983. An estimate was made, as of that date, as to whether repayment of an account was doubtful and if so, the amount which should be recorded as a reserve for the fiscal year ending November 30, 1982. The February date was chosen not only because of its proximity to the date when financial statements and tax returns were prepared, but also because it approximates the end of a natural business cycle in the men's retail clothing industry. By that time of the year, fall and winter clothing (shipped and sold by the plaintiff to its retail customers in the summer and fall of the previous year) had usually been sold by the retailers; retailers who had unsold fall and winter inventory as of February of a given year would carry that inventory over to the following fall season.

The plaintiff assessed its accounts receivable by first looking at the age of its overdue accounts. It identified all those which had been owing as of November 27, 1982 and which were still owing as of February 25, 1983. In the plaintiff's experience there was a significant co-relation between a retail customer's inability to pay for fall-winter goods by the end of February and the likelihood that the customer was in financial difficulty. In 1983 this led to the identification of 145 accounts.

The plaintiff allowed its customers credit terms of 30 days from shipment for made to measure items (which comprised five per cent to ten per cent of the plaintiff's business) and 60 days from shipment for all other items. Thus, by February 25, 1983, the goods covered by the 145 overdue accounts would have been shipped, in the case of made to measure goods no later than 120 days earlier, and in the case of all other goods, no later than 150 days earlier.

After identifying the 145 accounts, the senior officers of the plaintiff (Mr. McWhinnie and Mr. Enkin) reviewed each individual account to determine whether, in their view, there was a real risk that the account would not be paid and if so, what amount should be included in the plaintiff's financial statements as a reserve therefor. In determining the degree of risk, the factors taken into account included the age of the overdue account, the customer's financial position including debts owed to the bank and others, the history if any, of N.S.F cheques on the account, whether the customer's sales were increasing or decreasing, the personal characteristics of the customer such as health, work ethic, ability to retain staff, the particular local conditions pertinent to the customer and whether amounts had been paid on the account since November 27, 1982. The plaintiff had credit files on each customer, in which it kept copies of the customer's financial statements, any credit reports that might exist (such as Dun and Bradstreet reports), reports on visits to the customer which had been made by Mr. Enkin, Mr. McWhinnie or others and information which might have been gleaned from the plaintiff's salesmen. In February of 1983, the assessment of the accounts receivable was done in the context of the recession which existed at that time. The recession started in 1980 and was particularly severe in the west where a number of the plaintiff's customers were located. Mr. McWhinnie's evidence was that in recessionary times men's clothing stores are particularly hard hit and that in 1981 in the west, sales in speciality shops dropped in some cases as much as 40 to 50 per cent. All 145 accounts which had been identified were determined to fall within the doubtful category.

In estimating the amount which should be allocated as a reserve for doubtful accounts, the plaintiff took into consideration the factors listed above as well as the amount the plaintiff might hope to realize (cents on the dollar) if the customer went bankrupt (referred to as the break-up value of the account). As has been noted, it took into account any amounts which had been paid by the customer between November 27, 1982 and February 25, 1983 ('since payments"). It also took into account whether additional amounts had become due from that particular customer between November 27,1982 and February 25, 1983. The estimate of the reserve for each customer was a judgment call on the part of Mr. Enkin and Mr. McWhinnie. There was no precise formula applied.

Despite the fact that a customer was significantly in arrears the plaintiff, in general, continued to ship goods to that customer. This was true even in the case of next season's goods. The next season's goods had usually been specifically ordered by the customer some time previously and could not easily be placed elsewhere. More importantly however, since in many instances, the plaintiff was the customer's principal supplier, if further goods were not shipped the customer would with certainty be put out of business. Thus, continued shipments kept the possibility of repayment of the doubtful account alive. In addition, the business philosophy of Mr. Enkin, who had built the Coppley Noyes business since its purchase by the Enkin family in the early 1950s, was to extend extensive credit to fledgling or struggling retail stores. His business policy was to provide financial as well as business advice and assistance to these stores. It was Mr. Enkin's view that by sharing the business risk with these trade customers, he would encourage the development of retail businesses which would provide an increasing market for the plaintiff's products. This policy, while it would appear to have been a successful business policy, meant that the plaintiff extended credit to its customers far beyond what might normally be considered prudent.

In any event, the plaintiff's practice of assessing its November 1982 reserve for accounts receivable by reference not only to the payments received between November 27, 1982 and February 25, 1983 but also to amounts which had subsequently become overdue during that period meant that the amount of the reserve thus determined exceeded what it would have been, had only the former been taken into account. The amount outstanding on the doubtful accounts if payments thereto, but not debits, are taken into account is referred to in the evidence as "the amount outstanding on fall goods as of February 25, 1983”. I will refer to it in this manner for the purpose of these reasons.

The characterization of the amount owing as of February 25, 1983, as an amount outstanding on fall goods, is based on the assumption that all payments made by the plaintiff's customers between November 29, 1982 and February 25, 1983 were intended to pay off pre-November 27,1982 debts and not amounts which had become overdue since that date. When neither the debtor nor creditor designates that a payment is to be applied to a specific invoice, it is assumed that the payment is being made to pay off the earliest debt: Main Plumbing & Heating Supplies (Western) Ltd. v. High Park Investment Corp. (1985), 37 Alta. L.R. (2d) 394 (Alta Q.B.); C.R.B. Dunlop, Creditor- Debtor Law in Canada (Carswell, 1981) at 24-27.

The evidence establishes that the practice of estimating a reserve by reference not only to amounts paid after year end but also to amounts which become due after that date, is in accordance with generally accepted accounting principles. Dr. Hanna's evidence is particularly convincing in this regard. I quote:

2. Is managements policy of continued selling to past due accounts reasonable in the circumstances and what implications does this policy have for estimating a reasonable allowance for doubtful accounts at the fiscal year end of the taxpayer?

Clearly, the policy of continued selling to past due accounts can be (but need not be) an effective policy that can lead to enhanced profitability:

. . .the managerial objective is not to minimize doubtful accounts expense but to maximize net income. Too stringent a credit policy may cause loss of sales volume which more than offsets the reduction in the doubtful accounts expense.

[Mosich et al.: p. 324]

The implications of such a policy on the proper estimation of allowance for doubtful accounts are acknowledged both by the CICA Handbook:

All businesses, however, do not stop selling to a debtor as soon as there is a possibility, or even a probability, of loss. In some cases, to do so would force the customer into immediate bankruptcy with the resulting loss of the balance presently outstanding; whereas a continuation of credit on a restricted basis over a period of a year or more may give the customer an opportunity of restoring his financial solvency or, if not, of reducing the balance owing on the account and so reducing the loss. For this or other reasons, many businesses continue to sell accounts on which there is a known probability of loss, and in such cases, the allowance for doubtful accounts should include some provision against the probable ultimate loss on such accounts even though the actual balance outstanding at the end of the period may subsequently be paid.

[Section 3020.08, emphasis added]

and by other authorities:

If a customer is in financial difficulties it becomes difficult for a major supplier to decide whether credit should continue to be extended. If it is cut off, there is a real danger that the customer will go bankrupt through lack of supplies, so at times there is a tendency to continue to supply limited additional credit as a means of delaying or perhaps preventing the bankruptcy. . . . In this situation aging the accounts is not as indicative of their conditions as it normally would be.

[Crandall and Cohrs: p. 177]

While authorities most often tend to associate these circumstances with cases where the objective is to keep the customer in business only long enough to achieve a maximum recovery of balances receivable, the case of the taxpayer—an intentional policy of long run credit extension while accepting a higher: than normal risk—is acknowledged. In particular, the CICA Handbook requires that the allowance for doubtful accounts should include some provision for loss in such circumstances—even when the". . .actual balance outstanding at the end of the period may subsequently be paid.”

In my opinion, the taxpayer acted properly in making provision for doubtful accounts at the November/82 fiscal year end even in cases where actual balances had been since paid—either at February or September/83 or at any other time—if there was a risk that arose in the fiscal year ending November/82 relating to credit sales which were included in that years revenues. A proper determination of income and matching of revenues with expenses for a period requires that revenues be reduced where there is a". . .known probability of loss. . [14 .

[References are to Mosich, Larson, Lam and Johnston, Intermediate Accounting, 5th Canadian Edition, McGraw-Hill Ryerson, Ch. 7, 1988 and Crandall and Cohrs, Intermediate Accounting: An Analytical Approach, Prentice-Hall Canada, Ch. 5, 1986.]

I should note, as well, that the plaintiff's practice with respect to bad debt write-offs was to carry debts as receivables long after many businesses would have classified them as bad. The claimant was not quick to place accounts receivable in the bad debt category.

The plaintiff's procedure for assessing its reserve for doubtful debts for the 1982 taxation year was no different from that which it had followed in preceding years. The defendant had reassessed the plaintiff with respect to its 1976 and 1979 taxation years. The plaintiff appealed those reassessments and the dispute was settled in 1980 with the Minister agreeing that a reserve of $1,000,000 was appropriate for the 1976 year and $1,250,000 was appropriate for the 1979 taxation year. The taxpayer had claimed $1,533,800 and $2,003,000 respectively for those years; the latter amounts appear as reserves for doubtful accounts in the plaintiff's audited financial statements.

The amount agreed upon by the Minister and the plaintiff for the 1979 taxation year was $753,000 less than the amount which appeared in the taxpayer's audited financial statements for that year. The taxpayer reduced the reserves claimed for tax purposes each year subsequent to 1979, to an amount which was also $753,000 below the reserve appearing in its financial statements. Thus, for the 1982 taxation year, while the amount for doubtful debts which appeared in the plaintiff's audited financial statements and which is claimed in this litigation was $3,084,000, the amount which was included by the taxpayer in its 1982 tax return was $2,331,000.

Minister's Assessment

The defendant, through the instrumentality of the Minister of National Revenue, reassessed the plaintiff for its 1982 taxation year and allowed $1,400,000 for doubtful debts. As counsel for the Minister argues, the burden is on the taxpayer to disprove not only the Minister's assumptions (one of which, in this case, is that $1,400,000 is a reasonable allowance) but also to prove that the plaintiff's estimate is reasonable.

There seems little doubt that if GAAP apply to a determination of the amount of the reserve in question, then, the taxpayer has met both burdens. Mr. Luciani, the Department of National Revenue auditor started, first of all, with a comparison of the plaintiff's bad debt history: the amounts written off in a year as compared to the accounts receivable at the beginning of that year. It was his view that this type of comparison, over a period of years, would give a ratio which could be applied to the 1982 accounts receivable for the purpose of determining what would be a reasonable reserve for doubtful debts. This comparison was done for a six-year period; a ratio was obtained and applied to the 1982 accounts. This resulted in a figure of $285,000 being determined. The defendant's auditor regarded this as too low and abandoned that approach.

Dr. Hanna's evidence regarding the accepted procedures for estimating doubtful debts was:

Authorities are in general agreement on the proper procedures for the estimation of a firm's allowance for doubtful accounts. Essentially, procedures fall into one or more of four categories:

i) Estimation of the allowance as a percentage of sales, usually credit sales, for the period.

ii) Estimation of the allowance as a percentage of accounts receivable at the end of the period.

iii) Estimation of the allowance on the basis of different percentages applied to different age categories of accounts receivable.

iv) Estimation of the allowance on the basis of a detailed customer by customer analysis.

For companies with large numbers of small accounts receivable, methods 1 and 2 are often employed. Method 3 is more costly but is generally acknowledged to lead to a more accurate approximation—especially if accompanied by the use of supplemental information. Method 4 is the most costly but, carefully done, should lead to a more effective estimate of the proper amount of allowance in a particular case.

Any one or combination of these procedures should:

. . .recognize changes in credit policy, changes in economic conditions, or any other factor which might affect [the customers] ability to pay their debts.

[Welsch et al.: p. 196]

In my opinion, the procedures employed by the taxpayer, essentially method 4, were not only appropriate but represented the methodology to be employed if an accurate estimate of the allowance for doubtful accounts was to be achieved. The taxpayer's detailed review of customer accounts included the following desirable features: . . .

[Reference is to Welsch, Zlatkovich, Harrison, Nelson, Zin, Intermediate Accounting, 4th Canadian ed., Irwin, 1986.]

I would note that the reserve for doubtful debts claimed by the taxpayer had historically been much much larger than its bad debt write-off experience in the following tax year. For example, the reserve claimed in the 1981 taxation year was $2,235,000; the plaintiff's bad debt write-offs in 1982 were $510,840. The taxpayer's bad debt write-offs in 1983 were not quantitatively different than they had been in 1982. It was this difference between the plaintiff's bad debt history and the amounts included as a reserve for doubtful debts which prompted Mr. Luciani to review the plaintiff's 1982 reserve.

Mr. Hanna was asked about the significance of bad debt history in estimating a reserve for doubtful debts:

3. Do I agree that a ratio of “bad debts written off to accounts receivable” that is considerably below the ratio of “allowance for doubtful accounts to accounts receivable”, indicates a likely overallowance for doubtful accounts?

Generally, I would agree that this would seem to be the case. However, giver: the taxpayer’s policy of continued sales to past due accounts, so simple a conclusion cannot be drawn in this case. The appropriate method of estimating the allowance for doubtful accounts, in the taxpayer's case, must provide for both:

i) the portion of November/82 receivables that will not be collected, and

ii) the eventual probable loss that will occur because of the "continued sales to past due accounts” policy followed during and subsequent to the 1982 fiscal year. To the extent fiscal 1982 revenues have benefited from this policy, the cost of this benefit should be a charge against these revenues. While this applies to the 1982 fiscal year in this case, the concept applies to all years and some of the riskiness in accounts receivable at any date should have been charged against each of the years that the policy was in effect.

In my opinion, it is reasonable to observe that actual bad debt write off rates are considerably below doubtful account provision rates in the taxpayer's case so long as the"continued sales to past due accounts” policy remains in effect and where a detailed analysis of outstanding accounts receivable indicates a material risk of account collection.

In any event, the defendant's auditor, having abandoned the idea of estimating a reasonable amount for doubtful debts on the basis of the ratio described above, turned to a review of the actual accounts the taxpayer had identified in February 1983 as being doubtful. Mr. Luciani's review was done in September 1983. He took into account any payments which had been made on the respective accounts after February 25, 1983 and up to September 30, 1983, as well as those made between November 27,1982 and February 25, 1983. The former was, of course, information not available to the taxpayer when it prepared its financial statements and tax returns in the previous March and April.

Mr. Luciani was strongly influenced by the fact that the plaintiff had shipped new goods to the customers despite the fact that those accounts had been classified as doubtful by the plaintiff. Mr. Luciani considered this to be evidence that the plaintiff expected those accounts to be paid. In his view, when new goods had been shipped to customers, amounts with respect to those accounts should not have been included in computing the reserve for doubtful debts.

In preparing his estimate, Mr. Luciani looked primarily at whether there had been continued trading activity on the account and whether the customer had paid any further amounts to the plaintiff up to September 30, 1983. If the amount owing on fall goods as of November 27,1982 had been paid off as of September 30, 1983, no reserve was allowed for that account regardless of whether additional amounts had become due on the account during the period. If an account was still active as of September 30, 1983 and the amount owing on fall goods as of November 27,1982 had not been paid off, a reserve was allowed for the amount which was still owing. If, in Mr. Luciani's view, it was Clear that the customer was unable to pay his account as of September 30, 1983, then, an amount somewhat higher than the unpaid portion of the November 1982 receivable was allowed.

Mr. Luciani's review of the plaintiff's accounts was admittedly subjective (as any such estimate must be). He determined that $1,368,000 was a reasonable reserve for tax purposes. This was subsequently rounded up to $1,400,000. Mr. Luciani admits that his assessment was not based on the same knowledge of the accounts that the plaintiff had and that he did not consider factors such as the break up value of the account to the plaintiff. He did not talk to the officers of the taxpayer to ascertain why had they had included reserves with respect to certain accounts even though he admitted that the taxpayer's officers would be in the best position to evaluate the risk of non collection associated with an account. He had no knowledge of the particular business in which the plaintiff was engaged. He admitted that the amount outstanding on fall goods as of February 25, 1983 was of doubtful collectibility. Mr. Luciani's view is that the estimating of a reserve for tax purposes differs from the estimating of a reserve for financial statements purposes.

Legal Principles Applicable

The general principles applicable to the calculation of income for tax purposes, and to the interpretation of the Income Tax Act, are that they should be consonant with ordinary commercial and accounting principles and practices unless the Income Tax Act requires otherwise. In The Bank of Nova Scotia v. The Queen, [1980] C.T.C. 57; 80 D.T.C. 6009 at 62 (D.T.C. 6013), Mr. Justice Addy explained this principle as follows:

Generally recognized accounting and commercial principles and practices are to be applied to all matters of commercial and taxation accounting unless there is something in the taxing statute which precludes them from coming into play. The legislator when dealing with financial and commercial matters in any enactment, including of course a taxing statute, is to be presumed at law to be aware of the general financial and commercial principles which are relevant to the subjectmatter covered by the legislation. The Act pertains to business and financial matters and is addressed to the general public. It follows that where no particular mention is made as to any variation from common ordinary practice or where the attainment of the objects of the legislation does not necessarily require such variation, then common practice and generally recognized accounting and commercial principles and terminology must be deemed to apply.

See also The Queen v. Metropolitan Properties Co., [1985] 1 C.T.C. 169; 85 D.T.C. 5128, especially at 180-81 (D.T.C. 5137).

Thus, in determining a reserve for doubtful debts, the principles and factors that are used for the preparation of financial statements, as governed by the generally accepted accounting principles approved by the Canadian Institute of Chartered Accountants, are applicable, unless: (1) the Income Tax Act expressly requires otherwise or (2) the Income Tax Act implicitly requires otherwise. There is no definition of doubtful debts or other express provision in the Income Tax Act which requires a departure from GAAP in the circumstances of the present case. If a departure is required that result must arise because of a conflict of GAAP with other provisions or with the overall intent of the Income Tax Act.

Both counsel agree that the senior management of a corporate taxpayer is in the best position to determine, from its inspection of the company's accounts receivable, which accounts are likely to give rise to difficulty and might be of doubtful collection: Atlas Steels Ltd. v. M.N.R. (1961), 27 Tax A.B.C. 331; 61 D.T.C. 547 (T.A.B.) at 334 (D.T.C. 550) and Kenora Miner and News Ltd. v. M.N.R., [1970] Tax A.B.C. 337; 70 D.T.C. 1228 (T.A.B.). As counsel for the defendant stated, if this were not the case, the company would be in a sorry state indeed.

The jurisprudence which exists with respect to estimating reserves for doubtful debts, for tax purposes, indicates that delay in payment alone is not sufficient to justify including an amount in a reserve: No. 409 v. M.N.R. (1957), 16 Tax A.B.C. 409; 57 D.T.C. 136 (T.A.B.). Among the factors which may be taken into consideration in estimating a reserve are the time element (the age of the overdue account), the history of the account, the financial position of the client, any increase or decrease in the client's total sales, the taxpayer's past bad debt experience, the general business condition in the country and the business condition in the particular locality: No. 81 v. M.N.R. (1953), 8 Tax A.B.C. 82; 53 D.T.C. 98 (T.A.B.).

It is conceded that in order for an amount to be included as a reserve for doubtful debts there has to be more than just some doubt that the account might not be paid: Picadilly Hotels Ltd. v. The Queen, [1978] C.T.C. 658; 78 D.T.C. 6444 (F.C.T.D.). The decision in No. 87 v. M.N.R., supra, rejected the assertion that every debt which is overdue is a doubtful one against which a reserve must be set up; see also Brignall v. M.N.R. (1961), 27 Tax A.B.C. 233; 61 D.T.C. 488 (T.A.B.). There must be good and substantial reason to question the likelihood that the account will be paid. The Interpretation Bulletin issued by the Minister of National Revenue (No. IT-442, paragraph 22) describes the test as follows:

For a debt to be classed as a bad debt there must be evidence that it has in fact become uncollectible. For a debt to be included in a reserve for doubtful debts it is sufficient that there be reasonable doubt about the collectibility of it. . . .

In Highfield Corporation Ltd. v. M.N.R., [1982] C.T.C. 2812; 82 D.T.C. 1835 (T.A.B.) at 2828 (D.T.C. 1847), it was said:

A“ Reserve for doubtful debts" established under section 20(1)(l) of the Act would seem to leave with the taxpayer a much greater degree of flexibility in using business judgment with regard to the inclusion of amounts in such a reserve that is permitted to a taxpayer in claiming a deduction under section 20(1)(p) of the Act for a "bad debt”. The term "doubtful debt" in itself can mean only what it says— the debt is owing and possible of collection, but that possibility is not sufficiently certain in the mind of the taxpayer that he wishes to be placed in the disadvantageous position of having to pay income tax thereon before that possibility has become more of a certainty.

If there is a reasonable doubt that an account is not collectible, the degree of doubt is expressed as a proportion of the total debt taken as a reserve. In that sense the amount included in a reserve with respect to any given account is an estimate of the risk that the account will not ultimately be paid.

In MacDonald Engineering Projects Ltd. v. M.N.R., [1987] 2 C.T.C. 2237; 87 D.T.C. 545 (T.C.C.), the treatment of amounts paid on doubtful debts after year end but before a taxpayer filed his income tax return was in issue. The D.T.C. headnote describes the decision:

The Court found that the sum of approximately $35,600, which was actually paid before the taxpayer filed its tax return for the 1982 taxation year, could not reasonably be regarded as forming part of the reserve. However, in view of the general economic climate that existed at the time and in view of the history of the customer's accounts, the balance of the debt was properly regarded as doubtful.

This is consonant with normal accounting procedures which require that in preparing financial statements all current information be considered.

There is no jurisprudence dealing directly with the type of circumstances at issue in this case.

Defendant's Representation

The defendant's main contention is that what constitutes a reasonable reserve for doubtful debts for tax purposes differs from what is considered to be a reasonable amount for financial statement purposes. That is, while Dr. Hanna indicated that for financial statement purposes two factors should enter into the assessment, namely, the portion of the November 28, 1982 receivables that will not be collected and the eventual probable loss that will occur as a result of the policy of continued selling, it is argued that for tax purposes only the first is applicable.

There was some suggestion in the presentation of this case that the concept ” reserve for doubtful debts" found in paragraph 20(1)(l) of the Act is different from the concept ” reserve for doubtful accounts" used in the preparation of financial statements, because the terminology in which each is expressed differs. I do not think that conclusion can stand. In fact, counsel for the defendant did not place much reliance on it. If the concepts, for financial statement and tax purposes are different, then, the rationale must be found elsewhere; it cannot be based on the difference between the words "doubtful debts” and "doubtful accounts”.

It is argued that an estimate for a reserve for doubtful debts for tax purposes, pursuant to paragraph 20(1)(l) of the Income Tax Act, is not the same as the estimate of a reserve for doubtful accounts for financial statement purposes because the reserve claimed pursuant to paragraph 20(1)(I) is a reduction of the amount which the taxpayer has already included in income. It is argued that for reasons similar to those given in M.N.R. v. Anaconda Brass Ltd., [1955] C.T.C. 311; 55 D.T.C. 1220 (P.C.), generally accepted accounting principles should not be followed in the present case. It is contended that, as in the Anaconda case, the taxpayer's method of accounting creates a hidden reserve for use in future years.

With respect to the analogy which it is sought to draw, to the Anaconda case, supra, I think it is well known that the reasoning of the Privy Council in overruling both the Exchequer Court and the Supreme Court of Canada has been widely criticized. See, for example, Glassford, LIFO—Cost of Inventory under the Income Tax Act, 1 Osgoode Hall L.J. 61 (1959) and S.D. Thorn, Anaconda Comment, 4 Can. Tax J. 8 (1956). As I understand it, most commentators have been of the view that the choice of LIFO (last in first out) as a method of inventory costing is not in conflict with the purpose of the Income Tax Act providing that method is used consistently over the years by the taxpayer. Indeed, in the particular circumstances in which it is appropriate to use LIFO, that method gives a better assessment of the taxpayer's profits for the year than does FIFO.

The fact situation with which the Privy Council had to deal in the Anaconda case, supra, Was one where the taxpayer had changed its method of inventory costing from FIFO (first in first out) to LIFO. Both the decision of the Privy Council and that of the Supreme Court refer to the fact that the new LIFO method had been permitted in the United States for tax purposes but in the context of statutory safeguards ([1955] C.T.C. at 321 (D.T.C. 1225) (P.C.) and [1954] C.T.C. 335; 55 D.T.C. 1179 (S.C.) at 344 (D.T.C. 1183) (Kerwin, C.J. dissenting) and at 345 and 352 (D.T.C. 1183 and 1187) (Esty, J. dissenting)). There was no such legislation in Canada. This was undoubtedly a significant factor in the decision.

In any event, the Privy Council held that the application of LIFO to inventory costing meant that the taxpayer was setting up a reserve, as I understand the decision, which would never be used unless the taxpayer was going bankrupt or going out of business. Only when a taxpayer ceased purchasing new inventory would the original cost inventory be brought into the calculation of income. As I understand the criticism of the Anaconda decision, it is that such reasoning impliedly assumes a certain physical flow of inventory (which all decisions expressly disavow) rather than focusing on the question: which accounting method produces the best estimate of the taxpayer's real profits for the year. Consequently, I do not find the analogy to the Anaconda case, supra, persuasive. In addition, regardless of the validity of the reasoning used in the Anaconda case, I have not been convinced that the taxpayer in the present case issetting up a reserve similar to that which was allegedly being created in that case. The reserve for doubtful debts does not fluctuate on the basis of the financial health or continued business activity of the taxpayer but on the economic viability of its clientele.

Another aspect of the Minister's argument relies on the fact that the amount of the reserve for doubtful debts which is claimed in one year, must be added to the taxpayer's income the following year and the reserve for doubtful debts calculated anew. This, it is said, is different from the treatment of doubtful accounts for financial statement purposes because such statements reflect only the increase or decrease from year to year of a reserve for doubtful accounts. I do not think that a substantial difference exists between the procedure used in preparing financial statements and that which is required under the Income Tax Act. While a balance sheet may only reflect increases or decreases in the reserve, in fact, the reserve is calculated anew each year for financial statement purposes as it is for tax purposes.

Counsel for the Minister argues that GAAP expressly contemplates that a different regime will be used for tax purposes than is used in preparing financial statements. Paragraph 3020.14 of CICA Handbook states:

The allowance [for doubtful debts] should be determined in accordance with generally accepted accounting principles regardless of how the allowance is determined for taxation purposes.

This admonition does not support the argument which it is sought to make. All that paragraph 3020.14 says is that the two reserves may differ but not that they must differ. This is entirely consistent with paragraph 24 of the Minister's Interpretation Bulletin IT-442 which provides that a taxpayer may claim an amount as a reserve that is less than the total amount which may be claimed and that that lesser amount will still be viewed as ” reasonable". The Minister's Interpretation Bulletins, of course, are not authorities. Reference to them is made in these reasons only for the purpose of setting out the position which the Minister has taken in those publications. I do not consider them to be determinative of the issues which they address.

It is argued that GAAP are not applicable for tax purposes, when estimating a reserve, in a case such as the present, because under those principles a reserve can be claimed for a contingency. It is argued that GAAP allows a reserve to be claimed with respect to an anticipated loss that might occur two, three or five years later while for tax purposes only losses which are expected to occur in the immediate future can be considered. As I understand this argument, it is that for tax purposes the expectation of loss has to be more immediate and carry a higher degree of probability than is the case under GAAP. In addition, it is argued that the bad debt history of the taxpayer, while it might have limited relevance for financial statement purposes, has much greater significance for tax purposes. It is argued that, in this case, the fact that the taxpayer continued to ship goods to outstanding accounts after year end is very significant and indicates that the taxpayer expected the amounts outstanding on those accounts to be paid.

I have some difficulty with these conclusions. While counsel argues that under GAAP an allowance can be claimed for a contingency, the Interpretation Bulletin IT-442 indicates that this is equally the purpose of paragraph 20(1)(l) :

21. Paragraph 20(1)(I), which authorizes a deduction in respect of doubtful debts of the kind described therein, is an exception to the general rule set out in paragraph 18(1)(e) that a deduction may not be claimed for losses that are contingent in nature. . . .

In Day & Ross v. The Queen, [1976] C.T.C. 707; 76 D.T.C. 6433 (F.C.T.D.), it was held that a "reserve" in paragraph 12(1)(e) of the pre-1972 Income Tax Act (now paragraph 18(1)(e)) denotes the setting aside of an amount to meet a contingency, an unascertainable and indefinite event which may or may not occur. I have been referred to no authority or convincing argument which leads me to conclude that the test for estimating a reserve for doubtful debts has to be more restrictive and carry a higher probability of risk than is the case in applying GAAP. I am not convinced that an assessment of the taxpayer's real profit for a year is more accurately determined by assessing the reserve claimed without reference to the second factor to which Dr. Hanna referred (the eventual probable loss that will occur because of the "continued sales to past due accounts”), than it is by reference to both factors which he considered should be taken into account in this taxpayer's case.

Also, I can not conclude from the evidence that the plaintiff's reserve was set up in anticipation of losses which it was thought would occur two, three, or five years down the road rather than during the 1983 year. The plaintiff simply did not know whether or when a doubtful debt would become a bad debt. In fact, given the recession which existed at that time (which was less severe in September 1983 when Mr. Luciani did his assessment than it was in February and March 1983) an earlier rather than a later date was more likely.

With respect to the conclusion, the defendant seeks to draw from the postyear-end shipment of goods to customers, I have no doubt that the plaintiff hoped the accounts would be paid. Also, there is no doubt that the taxpayer shipped further goods to the customers for the purpose of avoiding immediate loss and to keep the possibility of repayment of those accounts alive. I do not conclude, however, that the accounts should not have been classified as doubtful. Indeed, at least part of the test which Mr. Luciani applied seems more akin to the identification of bad debts rather than doubtful debts.

Two other arguments made by the defendant must also be considered: the plaintiff identified doubtful accounts solely by reference to slowness of payment and the jurisprudence indicates that slow payment does not equate to doubtful debts; secondly, accountants proceed on the basis of a principle of conservatism, when calculating profits, and this is not appropriate in calculating income for tax purposes. With respect to the first, I could not conclude that the plaintiff's assessment was made solely on the basis of the age of the accounts in question. The evidence establishes that other considerations also entered into the plaintiff's identification of the doubtful accounts. With respect to the argument based on the "conservatism principle”, this is too general a consideration to allow me to draw the kind of specific conclusion therefrom which the defendant would wish. In addition, as Dr. Hanna indicated, that principle is always constrained by the requirement that financial statements fairly reflect the financial position of the company.

Amount of the Reserve

What then of the reasonableness of the plaintiff's estimate of the amount identified as a reserve for doubtful debts? Even if the correct principles have been chosen, the Minister argues that these were incorrectly applied. In addition, it is argued that it is not sufficient to demonstrate that Mr. Luciani applied wrong principles in assessing what would be an appropriate amount as a reserve for doubtful debts, but that the actual amount which was determined $1,400,000 must also be proven to be unreasonable.

Counsel for the defendant argues that I cannot refer this case back for a reassessment without identifying the exact figure which should be used as a reserve for the purposes of calculating the taxpayer's 1982 reserve for doubtful debts. He argues that to do otherwise would only encourage further litigation.

The Minister challenges the plaintiff's estimate of the reserve, $3,084,000, on the ground that it exceeds the amount outstanding on fall goods, in the 145 accounts, as of February 25, 1983. The total amount owing on the 145 accounts as of November 27, 1982 had been $4,537,900. The amount owing on these same accounts as of February 25, 1983 was $3,373,053. If the amount owing, as of February 25, 1983 is calculated so as to include all payments made on the accounts subsequent to November 27, 1982 but exclude all additional amounts which had become owing during that period then the February 25, 1983 balance would be $2,695,122 (amount outstanding on fall goods as of February 25,1983).

The Minister argues that the estimated reserve must be adjusted downward from the amount it would have been at year end, as a result of amounts which have been paid by a customer between November 27, 1982 and February 25, 1983, but the reserve cannot be adjusted upward as a result of any additional amounts which became due during that same period. It is argued that this follows because the reserve allowed under paragraph 20(1)(I) is with respect to amounts which have already been included [a]s income. [1] The plaintiff does not dispute the fact that " since" payments must result in a diminution of the amount of the reserve which would otherwise have been claimed. It claims, however, that it is equally entitled to take into account the fact that further debts have arisen on those same accounts.

As I understand it, the reserve for doubtful debts under paragraph 20(1)(1) must be adjusted downward as a result of amounts paid after November 27, 1982 because accounting principles require that all current information be taken into account when preparing financial statements subsequent to year end. Equally, it seems to me subsequent information concerning additional overdue amounts with respect to those same accounts should also be considered, as it is for financial statement purposes. In my view, it would be inconsistent to require a taxpayer to take into account after acquired information of one kind, but not of another. Conceptually, the reserve for doubtful debts relates to accounts receivable as of year end. The after acquired information, as I understand it, is used to assess the degree of risk attached to the collection of those accounts. The assessment of the accounts which is done subsequent to year end is not a recalculation of the amount which was owing at year end, as of a later period of time but an estimate of the degree of risk attached to the collection of the account. On that basis, I cannot conclude that because the reserve exceeds the amount outstanding on fall goods as of February 25, 1983, that the reserve is unreasonable. In addition, a finding that the amount of the reserve must never exceed the amount outstanding as of the date of the post year end review would mean that the "cap" thereby imposed would vary depending upon whether financial statements and tax returns were filed immediately on year end or one, two, three or six months later.

The plaintiff's estimate, as has been noted is $3,084,000. Dr. Hanna gave evidence that the correct accounting principles and procedures had been used. He did not consider whether the estimate reached by the plaintiff was reasonable (for accounting purposes). Mr. Willson gave evidence that the correct accounting principles and procedures had been used and that in his view a reasonable amount for a reserve would be within the range of 2.6 million to 3.0 / 3.1 million. His precise estimate had been $2,862,000. There is some confusion in the evidence as to whether his estimate was based on a misunderstanding that the outstanding balance on fall goods was $3,675,121, as of February 25,1983, rather than $2,695,122. The aged trial balance at February 25, 1983 showed an amount due of $3,675,121. This would include not only the amounts on the doubtful accounts which had been due in November 27, 1982, but also amounts which had subsequently become due. Mr. Willson was asked whether his opinion of the reasonableness of the reserve would have been any different had he known that the amount due as of February 25, 1983 was $2,695,122 and not $3,675,121. He indicated that it might have been. He was not questioned further along these lines. He was not asked whether he had been misled by the shipment dates of the invoices or whether he was not aware of the plaintiff's credit policy or of the nature of the February 25, 1983 aged trial balance. I think it is fair, in the circumstances, to conclude that there was no misunderstanding on his part.

I turn then to the Minister's estimate. As has already been indicated, I cannot conclude that it was based on proper principles. Too heavy a burden was placed on the taxpayer by concluding that, because goods were shipped after year end, the accounts to which those goods related could not be doubtful. The wrong principle was applied in refusing to consider all post-year- end information (the increased amounts owing after November 27, 1982 as well as amounts that had been paid on those accounts) when estimating the risk associated with collecting those accounts. Also, in my view, I have a serious doubt as to whether it is appropriate to take into account, when reassessing a taxpayer in circumstances such as the present, information not available to the taxpayer at the time that estimate was made. At the very least, such after acquired information must be used sparingly. In the present case, the economic outlook was brighter by September 1983 than it had been in February and March.

Although, counsel may be right in saying that it is not enough to prove that the Minister's assessment proceeded on wrong principles or without taking into account all factors relevant to the method chosen, it is not open to a Court, in my view, to subjectively choose the Minister's assessment over that of the taxpayer's in the absence of an evidentiary basis establishing the appropriateness of the method the Minister used in reaching the assessment, or the appropriateness of some other method which would lead to approximately the same estimate. No direct evidence was called by the defendant in this case to support her position. Consequently, the Court is left with a situation in which, as between the taxpayer's and the Minister's estimates, the evidence leads to the conclusion that the taxpayer's estimate was appropriate.

Some further comment might be made about the use of information available to tax auditors which was not available to the taxpayer at the time of the filing of his or her return. In Hogan v. M.N.R. (1956), 15 Tax A.B.C. 1; 56 D.T.C. 183, the Tax Appeal Board dealt with a situation in which bad debts had subsequently become recoverable. The member of the Board hearing the case said, at 15-16 (D.T.C. 192):

. . . I am not unmindful of the fact that some $1,700 out of a total of approximately $3,200 written off by the appellant as bad debts has since been recovered by the new company, but I am of the opinion that this should not prejudicially affect my view of the honest determination which I believe the appellant made with respect to his bad debts in the light of all the known circumstances at the end of September 1953.

The determination made by the Minister as to what were bad debts in 1953 was made long afterwards, namely, some sixteen or eighteen months after the last day of September, 1953, when altogether new facts were available on which to base this determination. Notwithstanding the recent indications of the Courts, and indeed of this Board, that facts arising subsequent to the taxation year in question may be considered by the Courts and the taxation officials in reaching a conclusion on income tax matters—(the latest of which appears in the judgment of Ritchie, J., in Rosenblat v. M.N.R., [1956] Ex. C.R. 4 at p. 12; [1955] C.T.C. 323 [at 329-30; 55 D.T.C. 1205 at page 1208])—I am of the opinion that this is not the case in the present circumstances. . . .

The Hogan decision, supra, focused on paragraph 11(1)(f) of the Income Tax Act which directs the taxpayer to include as a deduction "the aggregate of debts owing . . . that are established by him to have become bad debts in the year". Consequently it was held that the legislation precluded facts being taken into account which could not have become known until many months afterwards and which would not have been foreseen by the taxpayer at the time his decision to classify the debt as bad was taken. The decision in Anderton and Halstead United v. Birrell, 16 T.C. 200 at page 209 was quoted:

What the statute requires, therefore, is an estimate to what extent a debt is bad, and this is for the purpose of a profit and loss account. Such an estimate is not a prophecy to be judged as to its truth by after events, but a valuation of an asset de praesenti upon an uncertain future to be judged as to soundness as an estimate upon the facts and probabilities.

These cases of course deal with bad debts and the Hogan decision, supra, specifically relied on the words of the statute" in the year". Nevertheless, these decisions render support to the proposition that in reviewing a taxpayer's estimate of a reserve for doubtful debts there is potential for a very unfair burden to be placed on the taxpayer when information is relied upon which was not available to the taxpayer at the time his or her estimate was made. I think it would be an error to say that such information can never be taken into account but, equally, I think a great deal of circumspection should be used in relying upon it.

There is one last argument concerning the amount of the reserve claimed which should be considered. On at least one of the 145 accounts expressly referred to in the evidence, the amount of the reserve claimed exceeded the amount overdue on that account as of November 27, 1982. Mr. McWhinnie acknowledged that this was not appropriate. He stated that this had occurred as a result of rounding up the amount owed. A review of the accounts indicates that this occurred on more than one account. It is clear that the reserve on individual accounts cannot exceed the amount owed at year end. At the same time, a review of the accounts indicates that the overall difference in the total estimate of the reserve created as a result of these errors is not a substantial or significant one.

Conclusion

I have not been persuaded in the circumstances of this case that the generally accepted accounting principles which are used for calculating doubtful account for financial statement purposes should not be followed for the purposes of paragraph 20(1)(l) of the Income Tax Act. There is no express requirement that a modification to those principles be adopted. There is no implied requirement. Indeed, in my view, the application of the principles is more consonant with the purposes of the Act than is a departure therefrom.

In the present case, generally accepted accounting principles were applied by the taxpayer to determine the amount of the reserve which was claimed. The evidence establishes that the actual numerical amount identified by the taxpayer falls within the range which is acceptable on the basis of the application of those principles. In my view, the amount is therefore reasonable for paragraph (20)(1)(I) purposes subject to the minor corrections being made with respect to the reserves for those individual accounts mentioned-above which exceeded the amount overdue as of year end.

Appeal allowed.