House of Lords portcullis
House of Lords
Session 2006 - 07
Publications on the Internet
PDF Print Version pdf icon

Judgments - Her Majesty's Revenue & Customs (Respondents) v. William Grant & Sons Distillers Limited (Appellants) (Scotland) and Small (Her Majesty's Inspector of Taxes) (Respondent) v. Mars UK Limited (Appellants) (Conjoined Appeals)


    SESSION 2006-07

    [2007] UKHL 15

    on appeal from: [2005]EWHC Civ 553





    Her Majesty's Revenue & Customs (Respondents)


    William Grant & Sons Distillers Limited (Appellants) (Scotland)


    Small (Her Majesty's Inspector of Taxes (Respondent)


    Mars UK Limited (Appellants)

    (Conjoined Appeals)

    Appellate Committee

    Lord Hoffmann

    Lord Hope of Craighead

    Lord Walker of Gestingthorpe

    Lord Mance

    Lord Neuberger of Abbotsbury



    William Grant & Sons

    Colin Tyre QC

    Graham Aaronson QC

    (Instructed by McGrigors, Edinburgh)

    Mars UK Limited

    Graham Aaronson QC

    (Instructed by Dorsey & Whitney)


    HM Revenue & Customs

    Colin Campbell QC

    Miss Jane Paterson

    (Instructed by Solicitor to HM Revenue & Customs, Edinburgh)

    Small (HM Inspector of Taxes)

    David Milne QC

    Rupert Baldry

    (Instructed by Solicitor to HM Revenue & Customs)

    Hearing dates:

    5, 6, 7 and 8 March 2007






Her Majesty's Revenue & Customs (Respondents) v. William Grant & Sons Distillers Limited (Appellants) (Scotland) and Small (Her Majesty's Inspector of Taxes) (Respondent) v. Mars UK Limited (Appellants) (Conjoined Appeals)

[2007] UKHL 15


My Lords,

    1.  The method of computing trading profits for the purposes of income and corporation tax has been settled for many years. First you compute the profits on a basis which gives a true and fair view of the taxpayer's profits or losses in the relevant period. Then you make any adjustments expressly required for tax purposes, such as adding back deductions which the taxing statute forbids. The classic formulation of this method is by Sir John Pennycuick V-C in Odeon Associated Theatres Ltd v Jones (1970) 48 TC 257, 272-273 and it has now been codified in section 42(1) of the Finance Act 1998:

    "For the purposes of Case I or II of Schedule D the profits of a trade, profession or vocation must be computed on an accounting basis which gives a true and fair view, subject to any adjustment required or authorised by law in computing profits for those purposes."

    2.  Although the requirement that the initial computation shall give a true and fair view involves the application of a legal standard, the courts are guided as to its content by the expert opinions of accountants as to what the best current accounting practice requires. The experts will in turn be guided by authoritative statements of accounting practice issued or adopted by the Accounting Standards Board, which are given statutory recognition by section 256 and paragraph 36A of Schedule 4 of the Companies Act 1985.

    3.  The dispute in these appeals concerns the computation of the trading profits of Mars UK Ltd ("Mars"), which makes confectionery and pet food, and William Grant & Sons Distillers Ltd ("Grant"), which makes Scotch whisky, in the years ending 28 December 1996 and 28 December 2002 respectively. There is no dispute that the profits stated in their accounts have been computed on a basis which gives a true and fair view. In each case, in accordance with current accounting standards, certain deductions have been made for the depreciation of fixed assets. But section 74(1)(f) of the Income and Corporation Taxes Act 1988 provides that in computing profits for tax purposes, no sum shall be deducted in respect of "any sum employed or intended to be employed as capital in…the trade". Although the language is by no means clear, this has always been taken to prohibit deductions for the depreciation of capital assets: see Robert Addie and Sons v Solicitor of Inland Revenue (1875) 2 R 431. Any sum which has been deducted for depreciation in the computation of profits must therefore be added back. The question is to identify which sums have been so deducted.

    4.  In order to find the answer it is necessary to examine the methodology employed by Mars and Grant in making their computations. This followed the relevant accounting standards. First, Statement of Standard Accounting Practice ("SSAP") 12, which was in force in 1996 when the Mars accounts were drawn up, states in paragraph 2 of its Explanatory Note:

    "Virtually all fixed assets have finite useful economic lives. In order for the financial statements to reflect properly all the costs of the enterprise it is necessary for there to be a charge against income in respect of the use of such assets. This charge is referred to as depreciation…"

    5.  Paragraph 16 of SSAP 12 states:

    "The accounting treatment in the profit and loss account should be consistent with that used in the balance sheet. Hence, the depreciation charge in the profit and loss account for the period should be based on the carrying amount of the asset in the balance sheet, whether historical cost or revalued amount. The whole of the depreciation charge should be reflected in the profit and loss account."

    6.  In 1999 SSAP 12 was replaced by Financial Reporting Standard ("FRS") 15 (Tangible fixed assets). Paragraph 77 clarified and refined the requirements of paragraph 16 of SSAP 12:

    "The depreciable amount of a tangible fixed asset should be allocated on a systematic basis over its useful economic life. … The depreciation charge for each period should be recognised as an expense in the profit and loss account unless it is permitted to be included in the carrying amount of another asset."

    7.  Pausing at that point, the effect of these standards is that the depreciation deducted in the profit and loss account for a given period should correspond with the depreciation shown in the balance sheet as having occurred over that period, "unless it is permitted to be included in the carrying amount of another asset". What is meant by this exception is best explained by reference to SSAP 9 (Stocks and long-term contracts), which dates back to 1975 and was adopted by the Accounting Standards Board. Explanatory note 1 sets out the relevant general principle:

    "The determination of profit for an accounting year requires the matching of costs with related revenues. The cost of unsold or unconsumed stocks will have been incurred in the expectation of future revenue, and when this will not arise until a later year it is appropriate to carry forward this cost to be matched with the revenue when it arises; the applicable concept is the matching of cost and revenue in the year in which the revenue arises rather than in the year in which the cost is incurred."

    8.  This fundamental principle is given effect by taking the revenue which has arisen in the relevant year and deducting from it only those costs which are attributable to those sales. These costs may have been incurred in the year in question, or they may have been incurred in earlier years and carried forward, in accordance with the general principle, to be matched with the related sales when they occur. The costs of stocks which remain unsold at the year end are not deducted for the purpose of computing the profit in that year but are carried forward to be matched against the revenue from their sale in future years.

    9.  SSAP 9 then specifies what should be regarded as the cost of stocks. Paragraph 17 says that it includes not only the cost of purchasing the materials but also the "costs of conversion". That is defined to include costs "specifically attributable to units of production" such as direct labour and expenses and also "production overheads" which are incurred "in respect of materials, labour or services for production, based on the normal level of activity, taking one year with another". Paragraph 20 specifically provides that such overheads should include the depreciation of assets "which relate to production".

    10.  SSAP 9 therefore demonstrates that when paragraph 77 of FRS 15 lays down a general requirement that the year's depreciation shown in the balance sheet should be deducted in that year's profit and loss account but makes an exception for a case in which depreciation is "permitted to be included in the carrying amount of another asset", that is intended to include carrying forward an appropriate part of the depreciation as part of the cost of stocks, to be deducted as and when the stocks are sold in a future year.

    11.  Both Mars and Grant prepared their accounts in accordance with these standards. They divided the depreciation which occurred during the year or was carried in the opening stock figure into two parts, which I shall call A and B. A was the depreciation in fixed assets which related to the production of goods sold during the year or in assets which were not used for production at all. B was the depreciation in fixed assets which related to production of unsold stocks. They deducted A from the year's revenue and carried B forward as part of the cost of unsold stocks.

    12.  Thus the Grant profit and loss account showed "Turnover" of £137,512,000 and "Cost of Sales" of £99,340,000. The latter figure did not include B. Note 4 to the accounts said that operating profit was stated "after charging depreciation". It gave the figure for total depreciation (A and B) and deducted the figure for depreciation "included within stock" (B). The Mars accounts were less explicit. Note 5 said that profit on ordinary activities before taxation was stated after charging "depreciation on tangible fixed assets" and then gave a figure for A and B together. But Note 10 said that depreciation of £3,039,000 (B) had been "included in the stock valuation" and there is no dispute that the figure for "Cost of Sales" in the profit and loss account included A but not B.

    13.  My Lords, so far there is no dispute between the parties. The expert accountants on both sides were agreed that this was the way the computations had been made and that the resulting statement of profits was in accordance with the standards and gave a true and fair view. And on these admitted facts, I should have thought that it was plain and obvious that, as only A has been deducted, section 74(1)(f) does not require B to be added back.

    14.  The Revenue submit, however, that whatever the methodology described by the accounting standards may be, the taxpayers must be deemed to have deducted both A and B and then added a sum equal to B back into profits in some other character which does not affect the deduction in respect of depreciation. The effect on the profit computation is the same as if B had not been deducted, but that is only because the value of B and the sum added back happen to be the same. The basis for this submission is that a deduction of anything less than the entire depreciation in the year would be contrary either to some fundamental principle of accounting or to the requirements of the Companies Act 1985.

    15.  Submissions that accounts must comply with fundamental principles of accounting additional to the best practice of accountants as embodied in the accounting standards have been made in other cases. But they have always been rejected. In Odeon Associated Theatres Ltd v Jones (1970) 48 TC 257 Sir John Pennycuick V-C (at p. 273) described the argument as "entirely novel". It was rejected by the Court of Appeal in Gallagher v Jones (Inspector of Taxes) [1994] Ch 107. In this case, it takes the form of saying that profits must be ascertained by taking all the revenue received during the year, deducting all the costs (including depreciation) incurred during the year and making an adjustment for the difference between opening and closing stocks, treating an increase in stock value as if it was revenue. That is certainly one way of computing profits and there are dicta in earlier cases which show that half a century ago and more, judges and accountants thought it was the normal and obvious way of computing profits. It may have been the only practical method when record-keeping was not sufficiently sophisticated to enable one to make a meaningful attribution of costs in one year to sales in some future year. It is not, however, the philosophy of SSAP 9, which permits the cost of unsold stock to be carried over into future years and set against future sales. In this exercise, relevant depreciation is simply another cost.

    16.  Then it is said that treating depreciation carried in stock as a cost excluded from the current year's computation and held back for a future year's computation is a category mistake. Stock is an asset which has a value and cannot be a cost. But that seems to me to confuse the role of stock in a balance sheet with its role in a profit and loss account. The balance sheet is a statement of assets and liabilities on a given date and in that statement, stock is indeed one of the assets. The profit and loss account, on the other hand, is concerned with revenue and costs, and in that context, the figure for stock represents a cost which SSAP 9 requires to be kept out of the computation of profit for the year but recorded to be carried over into the computation for a future year.

    17.  That, says the Revenue, is all very well when stock is valued at cost. Then the value of the stock in the balance sheet and the costs held over in the profit and loss account happen to coincide and the conceptual differences are concealed. But what if the value of the stock, on account, say, of obsolescence, falls below cost? Then it must be entered in the balance sheet at realisable value and a corresponding adjustment made to the profit and loss account. In such a case, the figure representing stock in the profit and loss computation cannot simply be regarded as a cost. It is something entirely different.

    18.  In my opinion there is no difficulty in accommodating this situation if one keeps in mind that the profit and loss account is concerned with revenue and expenses. A fall in the value of stock to below cost, although it involves no immediate outgoing or loss of income, is something which the principle of prudence requires should be treated as an expense and reflected in a deduction from that year's profit. There is no conceptual problem about recognising such a write-down as an immediate expense but carrying the cost of stock forward to be a future expense.

    19.  In support of these arguments, the Revenue placed considerable reliance upon the judgment of Lord Millett, sitting as a judge of the Court of Final Appeal in Hong Kong, in Commissioner of Inland Revenue v Secan Ltd (2000) 74 TC 1. The taxpayer company borrowed money to buy some land and build a block of flats. Construction took three years, during which no sales took place. The company's accounts submitted to the revenue for those years treated interest on the borrowed money as carried forward to be added to the cost of the flats and, there being no sales against which any costs could be set off, showed neither a profit nor a loss. This would be entirely in accordance with the principle of SSAP 9.

    20.  The company then changed its mind and claimed that its accounts had been drawn up on a false basis. Section 16 of the Hong Kong Inland Revenue Ordinance provides that in ascertaining the assessable profits for any year of assessment, "there shall be deducted all outgoings and expenses to the extent to which they are incurred during the basis period for that year of assessment … including … interest." Therefore, said the taxpayer, the carrying forward of any outgoings and expenses as contemplated by SSAP 9 is prohibited and all such outgoings should have been deducted in the year in which they were incurred. The claim was to rewrite the accounts to show that the interest payments gave rise to a substantial loss in each year.

    21.  It seems to me there are two alternative answers to this argument. The first is that section 16, on its true construction, does not prohibit the SSAP 9 approach and allows interest to be "capitalised", that is to say, carried forward and treated as part of the cost of future sales. The second is that, if section 16 is regarded as entrenching the old method of computation by which one deducts the year's outgoings from the year's revenue and then makes an adjustment for an increase in the value of stock or work in progress, then the taxpayer was right in saying that the accounts were drawn up on the wrong basis. Interest should have been deducted. But, by the same token, since the company's trade was building and selling flats, the interest would have been a cost of the stock or work in progress and, in the absence of evidence that net realisable value was lower, should have produced an increase in the value of the stock or work in progress which constituted a corresponding revenue item. Thus the outcome would have been the same. The taxpayer was not entitled to have it both ways and adopt the old method for deductions and the SSAP 9 method for revenue.

    22.  The difficulty I have is that, with respect to Lord Millett, I am unclear about which of these two answers he adopted. On the one hand, he said (at p 9) that section 16 did not prohibit the capitalisation of interest. That suggests the first answer. On the other hand, he said (at p 12) that there had been a deduction for interest in each of the first three years but that it had been set off against an increase in the value of the stock. The fact that the profit and loss account showed neither of these items was "merely a matter of presentation". That suggests the second answer. But the Revenue rely upon the case as deciding that when a cost item like interest or depreciation is carried forward as part of the value of stock or work in progress, it has nevertheless in some sense been deducted in the year in which the cost was incurred. That seems to me an impermissible and confusing mixture of two different systems of computation.

    23.  Finally, I come to the Companies Act 1985. In Schedule 4, paragraph 1(1)(b) provides that—

    "every profit and loss account of a company shall show the items listed in any one of the profit and loss account formats…set out [in section B]."

    24.  Section B offers 4 formats and both Mars and Grant adopted Format 1, which is by far the most commonly used. This requires the company to show, among other things, "1. Turnover 2. Cost of Sales and 3. Gross profit or loss." Note (14), in relation to "Cost of Sales" says that it shall be stated "after taking into account any necessary provisions for depreciation or diminution in value of assets". Paragraph 18 provides:

    "In the case of any fixed asset which has a limited useful economic life, the amount of -

    (a)  its purchase price or production cost; or

    (b)  where it is estimated that any such asset will have a residual value at the end of the period of its useful economic life, its purchase price or production cost less that estimated residual value;

    shall be reduced by provisions for depreciation calculated to write off that amount systematically over the period of the asset's useful economic life."

    25.  In my opinion there is nothing in any of these provisions which prevents depreciation (or any other cost) being deducted in a subsequent year if that is calculated to give a true and fair view of the profits. Note (14) says that Cost of Sales shall take into account depreciation but says nothing about how it should be taken into account. Likewise paragraph 18 says that the balance sheet value of an asset should be systematically reduced, but says nothing about when that reduction should be taken into the computation of profit. Obviously any reduction in balance sheet value for depreciation must at some time be reflected in a profit and loss account deduction, but there is nothing to say that they must coincide. This is not surprising when section 226 of the Companies Act makes it clear that the overriding requirement is that the accounts shall give a true and fair view.

    26.  It remains only to remind your Lordships of the way these appeals have come before the House. The Special Commissioners heard and allowed the appeals of both Mars and Grant against assessments which added back that part of the year's depreciation which I have described as B. The Revenue then appealed successfully against the Mars decision to Lightman J and against the Grant decision to the Court of Session. Both Mars and Grant therefore appeal to this House; in the case of Mars, directly from Lightman J under section 12 of the Administration of Justice Act 1969. For the reasons I have given, I think that Lightman J and the majority of the Extra Division were wrong to hold that B should have been added back and I agree with the lucid dissenting opinion of Lord Reed. I would allow both appeals and restore the decisions of the Special Commissioners.


My Lords,

    27.  I have had the advantage of reading in draft the speech of my noble and learned friend Lord Hoffmann. I agree with it, and for the reasons he gives I too would allow the appeals and restore the decisions of the Special Commissioners.

    28.  I am conscious of the wise words of Lord Deas in Addie v Solicitor of Inland Revenue (1875) 2 R 431, 433: "I think it is better not to run the risk of making any confusion in the grounds of judgment by adding anything to what your Lordship has said." But I am conscious too of the fact that there was an acute difference of opinion in the Extra Division of the Court of Session in Grant's case, as the opinions of all the judges in that court indicate: 2005 SLT 888. As we are differing from the majority, the addition of a few remarks directed to what was said in those opinions seems appropriate.

    29.  It is not easy to summarise the views of the judges of the Extra Division in a few words. But I think that the essential points on which they differed can be identified in the following paragraphs: Lord Penrose, paras 78-80; Lord Osborne, para 98; Lord Reed (dissenting), paras 127-128.

    30.  Lord Penrose said that the essential error in the taxpayer's approach to what section 74(1)(f) of the Income and Corporation Taxes Act 1988 requires was in treating book keeping for expense as the reality. It had ignored the injunction in the authorities to focus on the reason for making any adjustment to reflect the opening and closing stock and work in progress. The court would yield to accountants full authority to determine on generally accepted accounting principles and practices the amounts to be taken to represent the tangible assets so held. But it was for the court to say that such an amount must be taken into account in computing the full amount of the profits and gains arising in an accounting period, as a reflection of the cost incurred in acquiring or producing stock and work in progress or, if lower, the net realisable value of those assets.

    "The focus has been on the asset, and on the recognition of the need to credit against current expenses an amount to represent that asset if matching of revenue and expenditure is to be achieved."

The concept which he was seeking to emphasise was the crediting against expenses of a closing figure in the profit and loss account for unsold stock and work in progress, to match the revenue and expenditure arising in each accounting period. If the profit and loss account was analysed in this way, the gross amount of the depreciation was deductible.

    31.  For Lord Osborne the key to the issue lay in the direction in para 18 of Schedule 4 to the Companies Act 1985 that the purchase price or production cost, or that amount less any residual value, must be written off systematically over the period of the useful economic life of the asset. Depreciation was thus, as he saw it, temporarily associated with the useful economic life of the asset to which it related. But it seemed to him that, whenever a portion of the depreciation of a fixed asset in a particular year was used as an element in the calculation of the value of the closing stock, that element ceased to be properly capable of being regarded as depreciation. The element had become dissociated in a temporal sense from the useful life of the asset, and had become absorbed with all the other elements used to make up the homogeneity of the closing value of the stock. So that portion of depreciation, along with the remainder, required to be added back in the year in question.