Avon Products Pty Limited v Federal Commissioner of Taxation [2004] FCA 475

FEDERAL COURT OF AUSTRALIA

Avon Products Pty Limited v Federal Commissioner of Taxation [2004] FCA 475 SALES TAX – application for refund of overpaid sales tax – regular prices set in excess of costs including sales tax but otherwise without regard to amount of sales tax – all products regularly discounted - regular sales campaigns target overall profit margin on sales – whether applicant satisfied onus of showing sales tax not passed on to purchasers. Sales Tax Assessment Act 1992 (Cth) s 51, Sch I Taxation Administration Act 1953 (Cth) s 14ZZ Amway Australia Pty Ltd v Federal Commissioner of Taxation (1999) 41 ATR 443 Otto Australia Pty Ltd v Federal Commissioner of Taxation (1991) 28 FCR 477 Otto Australia Pty Ltd v Federal Commissioner of Taxation (1990) 25 FCR 257 Case 45/95 (1995) ATC 395 AVON PRODUCTS PTY LIMITED v FEDERAL COMMISSIONER OF TAXATION N978 of 2002 HILL J 23 APRIL 2004 SYDNEY

IN THE FEDERAL COURT OF AUSTRALIA

NEW SOUTH WALES DISTRICT REGISTRY N978 OF 2002

BETWEEN: AVON PRODUCTS PTY LIMITED

APPLICANT

AND: FEDERAL COMMISSIONER OF TAXATION

RESPONDENT

JUDGE: HILL J

DATE OF ORDER: 23 APRIL 2004

WHERE MADE: SYDNEY

THE COURT ORDERS THAT:

1. The application be dismissed.

2. The applicant pay the respondent's costs of the application.

Note: Settlement and entry of orders is dealt with in Order 36 of the Federal Court Rules.

IN THE FEDERAL COURT OF AUSTRALIA

NEW SOUTH WALES DISTRICT REGISTRY N978 OF 2002

BETWEEN: AVON PRODUCTS PTY LIMITED

APPLICANT

AND: FEDERAL COMMISSIONER OF TAXATION

RESPONDENT

JUDGE: HILL J

DATE: 23 APRIL 2004

PLACE: SYDNEY

REASONS FOR JUDGMENT 1 This is an application under s 14ZZ of the Taxation Administration Act 1953 (Cth). In it the applicant appeals against an objection decision of the Respondent Commissioner of Taxation disallowing to the applicant a claim for a credit entitlement under s 51(1) of the Sales Tax Assessment Act 1992 (Cth) ("the Act") 2 The applicant claimed a refund of sales tax pursuant to Credit Ground CR1 of Table 3 to Schedule 1 of the Act for overpaid sales tax during the period 1 March 1993 to 31 August 1998 ("the relevant period"). Credit Ground CR1 applies where tax has been overpaid. The ground is detailed in column 3 of the Table to be "Claimant has paid an amount as tax that was not legally payable". The amount of the credit (column 4) is stated to be "the amount overpaid, to the extent that the claimant has not passed it on". 3 It is not in dispute that the applicant overpaid sales tax in the amount of $3,610,261 during the relevant period. For completeness I will set out under the heading "The Facts" how the overpayment arose. What is in dispute here is whether the applicant passed the amount of the overpaid tax on to its customers. The Facts 4 During the relevant period, the applicant carried on a business which involved selling by retail a variety of goods that can be largely divided into two categories, namely cosmetics, fragrances and toiletries ("CFT products") which accounted for approximately 60% of total sales; and jewellery, accessories, gifts, apparel and home items ("non CFT products") which accounted for approximately 40% of total sales. 5 The applicant conducted its business through a network of sales representatives who were independent contractors, and not employees. The sales representatives sold the applicant's products directly to the public through door-to-door sales. Those sales were "indirect marketing sales" within the meaning of s 20 of the Act. In consequence the taxable value for the purpose of sales tax was the notional wholesale selling price under assessable dealing AD2d or AD12d. 6 Until 1 December 1995, the applicant had paid sales tax upon its indirect marketing sales on the basis that the taxable value was determined by reference to "cost plus 35%". This was in accordance with sales tax ruling ST2424. From 1 December 1995, the applicant utilised the Australian Taxation Office's ("ATO") "safe harbour" ruling of "cost plus 15%" (see Exhibit CJS31 to the Affidavit of Christopher Stevens). On 5 February 1999, the ATO issued a Sales Tax Private Binding Ruling to the applicant which determined that the taxable value of its products should be the "store cost of the goods plus 11.63%". Accordingly, the applicant claimed a refund of sales tax from the ATO because it had during the relevant period remitted sales tax to the ATO upon the basis of taxable values in excess of the "store cost of the goods plus 11.63%". 7 The question whether sales tax has been passed on may be a difficult hurdle for taxpayers seeking a refund to overcome. Whether or not this is the case, there have been few cases where the Court has had occasion to discuss the principles involved. Perhaps the reason for this may lie in the fact that the question is really one of fact and not one of law. No doubt a taxpayer which has charged sales tax at one rate and has thereafter found that the tax properly payable was a lesser amount could, if it wished, refund the amount of the overpayment to customers to whom it had sold goods and in such a case the taxpayer would obviously be entitled to a refund. However, that seldom happens in the real world. Indeed, it may often be impractical as customers may well have gone out of existence or be incapable of being located. That will particularly be the case where sales have been made to the public. In some cases taxpayers may be able to demonstrate that while not having given refunds, they have priced products in such a way as, in effect, to refund amounts previously overpaid to customers by reducing the price at which future goods are sold. Whether a taxpayer in such a case would be successful would depend upon the evidence which is adduced. In other cases a taxpayer may seek to show that it has priced its goods at a price that has the consequence that it is the taxpayer, and not the purchaser from the taxpayer, who bears the sales tax. That is what the applicant seeks to do in the present case. Accordingly it is necessary to examine carefully the evidence, particularly as to the applicant's methodology for setting prices, and whether or not changes in sales tax affected this methodology. 8 During the relevant period, the applicant sold all of its goods through marketing campaigns that were standardised across its 110 sales districts within Australia. There were 18 such campaigns during a calendar year and each had a duration of approximately three weeks. Although the applicant would set a "regular price" for each product sold, there was heavy discounting in each campaign. In fact, discounting was an integral part of the applicant's business, because 85% to 95% of its sales were made at a discounted price. Accordingly, to understand the applicant's methodology for price setting, it is necessary to look at how the applicant set both the regular price and the discount price for each of its products. To this end, I will first review the method for setting regular prices. 9 Prior to introducing a new product, the applicant would engage in "benchmarking" its prices against prices for comparable products charged by competitors. The benchmarked price became the regular price for the product. If there was not a directly comparable product, the applicant set its price at that of a similar product from within the Avon range, or if there was no such product, at the price which it thought the market could bear. 10 Evidence was given from Mr Christopher Stevens, who during the relevant period was initially Marketing Manager and later the applicant's Vice President of Sales. He explained the applicant's pricing of products by taking by way of example a range of bubble baths products available in 3 variants (Pink, Peach and Lavender). These products were introduced during the relevant period. 11 The first step in the pricing process was a product review meeting. For this meeting there was prepared for discussion a "Marketing Product Profile Sheet". On that sheet were recorded a proposed price ($9.95), a cost objective ($2.25) (ie an estimate of the amount which was the preferred maximum cost of the product), a gross profit margin based on the cost objective figure and the regular price net of the commission payable to sales representatives (67.7%). Sales tax was included in the "cost objective". It is said that the purpose of determining the objective cost of the product is so as to ensure that Avon achieved what to it was an acceptable profit margin on the sale of the product. It was not the case that Avon set prices specifically by adding a fixed profit margin to cost. 12 Next, the product profile was discussed at a New Product Development Meeting attended by a Product Manager with knowledge of costings from within the applicant's finance department and the representative from the purchasing and manufacturing area. This meeting would focus on matters such as the product design, quality, launch requirements and timing. Thereafter the representative from the purchasing and manufacturing area would send details of the product to a sample of vendors for a cost quote. 13 Once cost quote details were known, the Avon Finance department analysed the cost of the product. A cost sheet was prepared with a full breakdown of the estimated maximum cost of the product. The estimated cost could prove to be inaccurate as it might be affected by matters such as currency fluctuation, increase in supplier costs or other factors beyond the applicant's control. The cost sheet for the bubble bath products showed that each of the pink, peach and lavender varieties cost $2.137, $2.342 and $2.328 respectively. Sales tax of $0.489, $0.536 and $0.533 respectively were included in these figures. There was calculated a "net to Avon Price" which was the estimated retail price less the average sales representative's commission, which was approximately 30%. 14 There was then held a Cost Approval Meeting attended by the same people as attended the New Product Development Meeting. If the profit margin was below a minimum acceptable level the product was not introduced. If it was acceptable the product was considered for introduction. 15 In setting the regular retail price for the product, Mr Stephens said that the applicant considered both the prices of comparable products sold by its competitors and the price which it thought the market could bear. As noted earlier, the estimate of cost (which estimate included sales tax) was calculated to ensure that the ultimate sales of the product would produce an acceptable profit margin. 16 In cross-examination, Mr Stevens' agreed that cost was an important consideration to the applicant that was regularly reviewed. Consideration was given to reducing costs, for example by changing sourcing arrangements. It was accepted by him that the applicant always ensured that it covered its costs including sales tax. It was clear that Avon would not deliberately sell any product at a loss. 17 With non CFT products, the applicant conducted focus groups known as "Clinics". These groups involved the applicant's sales representatives. They were asked what price they expected to sell the product at and the quantity of the product they believed they would sell. The clinics assisted the applicant to determine the regular price where there was no comparable product already in the market. Otherwise the regular price for non CFT products was set by reference to the price of competitive products if there were any. 18 Regular prices were reviewed annually by the applicant's marketing department. Occasionally, this involved the applicant commissioning external market research companies to conduct confidential research among members of the public who purchased the applicant's products. 19 It is clear from the evidence that Avon seldom sold its products at the "regular price", because it regularly discounted the price of its products. 20 Before discussing how discount prices were determined, it should be said that the applicant sold all its goods through marketing or sales campaigns of which there were 18 in each calendar year, each with a duration of approximately 3 weeks. A brochure was prepared for each campaign and that brochure set out details of the products available for the period of the campaign and the prices at which they would be available. 21 Discount prices were said to be determined through a lengthy review process having regard to a number of factors. First, an overall target gross profit margin for each campaign was set at the applicant's Annual Planning Conference. The Marketing Department and Product Managers then held a Quarterly Planning Conference nine months prior to the relevant quarter's campaigns to review and revisit the plans and targets set at the Annual Conference. Following this, the Product Managers would develop detailed category plans for each campaign in that quarter, and ultimately the information for each category for a particular campaign would be consolidated by the relevant Campaign Planning Manager into a "Leader List". The Leader List set out all of the products in the campaign, the regular price, the offer price, the cost estimate, estimated number of units that would be sold, the estimated net revenue to the applicant, the estimate gross profit and margin on each product, as well as the total estimated net sales to the applicant and total gross profit and profit margin. It was this information that would be utilised to 'tweak' the discount prices so that target margins would be achieved. 22 The actual discount prices themselves were suggested by Product Managers based upon their previous experience and knowledge of the market and the price points that customers found attractive. The depth of the discount was driven by information from previous sales history, and in particular the price elasticity of the product. The applicant usually required that each individual CFT product be discounted at least once per quarter and approximately 85% to 95% of sales of CFT products were made at discount prices. It would seem in fact that if not all, then almost all products were discounted from time to time whether they were CFT products or not. Non CFT products were often "limited life products" that is to say they were generally made available only for two consecutive campaign of 3 weeks each and then discontinued. Obviously discounting did not arise with limited life products. 23 So far as the evidence shows, prices were never set by Avon on the basis of cost plus profit. But on the other hand, clearly prices were never discounted below a figure where profit would become loss. Discounts might take the form of a price set lower than a regular price or by the process of offering two or more product items for the price normally payable for one. 24 Ultimately however, the evidence demonstrated that in view of the fact that each campaign had a particular gross margin target as well as sales targets, the discounts were largely calculated to achieve the targets which were set in advance. The overall philosophy of the Avon sales campaigns in the relevant period was to have a mix of Avon products which not only provided a pleasing range of purchase options for its customers, but which also contained products having profit margins which, when aggregated, produced the desired overall campaign profit margin, number of units sold and sales. 25 This point was elaborated upon in Mr Stevens' cross-examination where he explained that if the projected sales for a campaign were too low, it could be given more "energy" by discounting a high margin product. That is, by lowering the price of a high margin product, more units of that product would be sold, with the result that the overall gross margin for the campaign was increased. 26 The applicant asserted that its pricing did not take account of sales tax, and that its prices were not affected by changes in sales tax rates. That is true to the extent I have already noted, that is to say that cost did inform the prices at which Avon sold its products, although not in the sense that prices were fixed on the basis that a fixed margin was added to cost. In the relevant period the rate of sales tax on Avon products increased on 18 August 1993 from 20% to 21%, and further increased on 1 July 1995 from 21% to 22%. From 1 December 1995, the applicant switched from "cost plus 35%" to "cost plus 15%" for calculating the taxable value of its products, but it is said that during the relevant period the applicant did not change the prices of its products to reflect this decrease. There is, however, a question what is meant by saying that pricing did not reflect the change of sales tax in the relevant period. 27 As I have endeavoured to make clear the regular price of CFT products can be said not to have changed at all in the relevant period. However, the regular price was not really the price at which CFT products were sold in that they were regularly discounted. The quantum of the discount depended upon the overall profit margin which it was calculated should be achieved by the particular sales campaign during which goods were sold. 28 The applicant tendered summary documents for the prices of its products during campaigns 13 to 18 in 1993, campaigns 9 to 18 in 1995 and campaigns 1 to 9 in 1996. Using the example of the bubble bath, it appears that the regular price for this product remained at $9.95 for the period spanning 1993 and 1996, and that it was intermittently discounted throughout the period to prices of $6.95 and $7.95. It is difficult to determine whether the same was true for all of the applicant's products. 29 Some reliance was placed by the applicant on the fact that two of the applicant's 110 sales districts comprised Norfolk Island and Christmas Island respectively. It is common ground that notwithstanding the fact that no sales tax was payable in either of those jurisdictions, the applicant did not alter its prices for products sold on those particular islands during the relevant period and that the prices prevailing there were the same as the prices prevailing in other sales territories where sales tax was in fact paid. The applicant argued that this was evidence of the fact that it did not pass on the overpaid sales tax to its customers in its prices because its prices were not set having regard to the amount of sales tax payable. 30 However for each campaign, the applicant prepared a brochure setting out details of the products available for purchase. Some 500,000 to 1,000,000 brochures were printed each campaign. Indeed printing costs were approximately $200,000 each campaign. Separate brochures were not printed for Norfolk Island and Christmas Island. Hence it was more a matter of convenience than any other factor which caused there to be no real difference between the prices applicable in either Norfolk Island or Christmas Island on the one hand and the prices applicable to other sales districts adjacent to those territories on the other. 31 Ultimately it was conceded, properly by Mr Stevens and, quite properly, that sales in Norfolk Island and Christmas Island were so small as to be insignificant to the applicant. I would therefore place no significance upon the circumstance that the prices in Norfolk Island and Christmas Island where no sales tax was payable were the same as those prevailing elsewhere in Australia.