AN ALTERNATIVE PRICING METHODOLOGY FOR NON-FOOD RETAIL

ERASMUS UNIVERSITEIT FACULTEIT BEDRIJFSKUNDE AN ALTERNATIVE PRICING METHODOLOGY FOR NON-FOOD RETAIL ANSGAR-JOHN BRENNINKMEIJER ERASMUS UNIVERSITEIT FACULTEIT BEDRIJFSKUNDE AN ALTERNATIVE PRICING METHODOLOGY FOR NON-FOOD RETAIL Afstudeerscriptie als onderdeel van het doctoraal bedrijfskunde, variant Logistiek, -februari 2001door Ansgar-John Brenninkmeijer Afstudeercommissie: -Prof. dr. S.L. van de Velde (Coach/voorzitter) -Dr. A.N.A.M. Boons (Meelezer 1) -Prof. dr. ir. G.H. van Bruggen (Meelezer 2) Cover illustration: "In the meantime, there were more fellow countrymen who tried their luck in selling and trading in the neighboring countries, and competition started to play a big role. There is a story that August went to the country fairs and offered two-penny stamps for one penny, in order to prove, that he had better offers than anybody else!"(Broekman, 1990) Het auteursrecht berust bij A.J. Brenninkmeijer. De inhoud is voor rekening van de auteur. De Faculteit Bedrijfskunde is slechts verantwoordelijk voor de onderwijskundige begeleiding en aanvaardt in geen enkele opzicht verantwoordelijkheid voor de inhoud. CONTENTS CKNOWLEDGMENT ......... ERROR! BOOKMARK NOT DEFINED.II SUMMARY................................................................................................ IV INTRODUCTION........................................................................................2 CHAPTER 1: INTRODUCTION TO RETAIL ACCOUNTING .........5 1.1 RETAILING AND THE VALUE CHAIN .......................................................5 1.2 CATEGORIZATION OF RETAILERS ..........................................................7 1.3 INCOME STATEMENT .............................................................................8 1.4 THE RETAIL INVENTORY METHOD .....................................................10 1.4 THE GROSS MARGIN AND PRICE SETTING ............................................12 1.5 STOCK TURNOVER AND VOLUME VERSUS MARGIN ............................16 CHAPTER 2 COST CALCULATION AND PRODUCTIVITY MEASUREMENTS ...................................................................................19 2.1 GROSS MARGIN RETURN ON INVENTORY INVESTMENT .......................19 2.2 GROSS MARGIN RETURN ON SPACE .....................................................21 2.3 GROSS MARGIN RETURN ON LABOR ....................................................23 2.4 THE TRINITY MODEL ...........................................................................26 2.5 DIRECT PRODUCT PROFITABILITY .......................................................26 2.5.1 Calculating DPP .........................................................................29 2.5.2 DPP in practice ..........................................................................31 2.5.3 Application of DPP in the non-food sector ................................32 CHAPTER 3 DESCRIPTION C&A FORMULA IN GERMANY .....35 3.1 COMPANY HISTORY .............................................................................35 3.2 STORE POSITIONING AND MERCHANDISING STRATEGY .....................36 3.3 IMPLEMENTING MERCHANDISE STRATEGY ........................................39 3.3.1 Organization for Merchandise Management .............................39 3.3.2 The buying function ....................................................................40 3.3.3 Merchandise Managers and the flow of goods ..........................42 3.3.4 Store operations ..........................................................................46 CHAPTER 4 UNIT COST RESEARCH AND THE ASSORTMENT MODEL .......................................................................................................47 4.1 COST ACCOUNTING RESEARCH METHOD C&A GERMANY ...............47 4.1.1 Factories as an image of store organization .............................47 4.1.2 Research Scope and Objectives..................................................50 4.1.3 Broad description of the research ..............................................52 4.2 ILLUSTRATION OF THE METHOD USING EXAMPLES .............................55 4.2.1 Warehouse costs .........................................................................55 18-6-09 ii 4.2.2 Transportation costs ...................................................................56 4.2.3 Store costs ...................................................................................57 4.2.4 Editing and presentation of the data ..........................................62 4.3 COMPARISON RESEARCH METHOD C&A AND CONVENTIONAL MEASURES .................................................................................................63 4.3 COMPARISON RESEARCH METHOD C&A AND DPP ............................65 4.4 THE ASSORTMENT MODEL .................................................................66 4.4.1 Density of units per square meter ..............................................68 CHAPTER 5 METERS, THE ALPHA AND OMEGA ........................71 5,1 CHANGING THE INDIRECT COST APPLICATION BASE ...........................71 5,2 COMPANY PLANNING WITHOUT AN AVERAGE GROSS MARGIN PERCENTAGE TARGET ................................................................................73 5,2,1 Company level planning .............................................................73 5,2,2 Business unit level planning .......................................................74 5,2,3 Style contribution ........................................................................75 5,3 USING THE ALTERNATIVE METHOD IN PRACTICE ................................76 5.5 STRENGTHS AND WEAKNESSES OF THE ALTERNATIVE METHOD .........79 CHAPTER 6 CONCLUSION ..................................................................80 BIBLIOGRAPHY ......................................................................................82 APPENDIX ...................................................................................................1 18-6-09 iii SUMMARY by Annerie Brenninkmeijer Traditionally retailers, especially non-food retailers such as clothing stores, set the prices of all the goods in their store by multiplying the cost of goods sold by one standard, company-wide (or department wide) gross margin percentage. This paper offers an alternative to this traditional fixed margin strategy, in the form of a variable gross margin strategy. This new pricing & planning strategy draws upon the learnings from two models used by food retailers (namely the Trinity Model and Direct Product Profitability or DPP) as well as two studies conducted for C&A in the 1980's and 1990's. These learnings make it possible for non-food retailers to gain more insight into the SGA (selling, general and administrative) expenses per item so that they can support competitive pricing decisions at a decentral level and still remain profitable. Traditionally non-food retailers focus on simple gross margin percentages per department or per store because the figures are accurate, easy to calculate and suited for benchmarking versus competition. In addition, non-food retailers pay a great deal of attention to stock turnover due to the success of low price-high turnover retail strategies (such as Wal-Mart and H&M). Thanks to the introduction of new models like the Trinity Model and DPP, food retailers (like Albert Heijn and Tesco) have been able to refine their profitability measures much further by also taking into account how they utilize their selling space as well as labor. Non-food retailers have resisted the introduction of similar models for the following reasons: (1) the larger and constantly changing assortments, which makes it more costly and complex to collect the necessary data. (2) The traditionally higher gross margins and the importance of fashion, so that non-food retailers prefer to focus on getting the right styles and colors into the store rather than on analyzing SGA expenses. (3) The less straightforward labor processes (i.e. store personnel does not just stock the shelves as in supermarkets. They also advise customers, man fitting rooms, etc.). (4) Finally, the lack of detailed (shelf) space plans. Two studies conducted for the clothing chain C&A offer potential solutions, particularly to issues 1, 3 and 4. Firstly both studies focussed on general clothing styles rather than on sizes and colors, 18-6-09 iv thus significantly reducing the number of individual products. Secondly, one study introduced the idea of Multi Moment Recording which entailed using a small sample of stores and a limited number of days to measure and record the activities of store personnel and cashdesk staff and thus allocate the costs of these activities to individual styles. Thirdly, the last study introduced the simple idea of space density (or the number of units per square meter) to include the impact of space utilization or style profitability. With the use of Multi Moment Recording, direct labor costs can be linked to specific styles. Through the use of the space density concept indirect store overhead costs can be allocated to styles on a square meter basis. Thus, a large part of the SGA expenses can be allocated to unique product lines. Individual department managers can then use this information to determine what the individual gross margins and prices need to be per style in order to cover the remaining expenses and the department profit target. In this scenario, the gross margin per product is variable. Company wide, fixed gross margin percentage targets are no longer required. (The system is described in more detail in Chapter 5). The challenge now is to get full top management support for the new system and to work out potential glitches by putting it into practice. 18-6-09 1 Introduction Background Over a period of two years, I worked as a men's shoe buyer for C&A Europe. During this time I started thinking about the traditional methods of price setting and cost calculation in retail. It seemed strange that retail accounting assumes a direct link between the cost of goods sold (merchandise cost) and selling, general and administrative (SGA) expenses (also known as operating expenses). Retail literature, especially books and articles on non-food retailing, are mostly concerned with aggregated margins and costs. In the past, few authors questioned the fact that retail prices are calculated by multiplying the product cost price by a standard percentage. During the last decades changes have occurred especially in the food sector (supermarkets) due to the introduction of information technology and the use of standard barcodes on products. Space management systems support the calculation and allocation of costs to individual products. Activity Based Costing in general and Direct Product Profitability (DPP) in particular are used by a great number of food retailers. The non-food sector however, lags behind in these developments. The constantly changing assortment and the way goods are presented, do not seem to be suited for the methods used in food retailing. Within C&A I came across two separate studies, one empirical and concerned with costs, the other theoretical and concerned with revenues. It seemed that combined together they might lead to a supplement or alternative for the standard pricing methodology as it exists today. Subject and scope The subject of this paper is the ways retail companies can calculate the costs connected with selling individual products (stock keeping units or styles). The scope of the research is limited to the operational costs of the retailer. The focus is on costs that occur in the actual stores. The production process from raw material to finished product as well as transport to the retailer's distribution center is not included. The empirical data is mainly limited to the stores of C&A Germany during the years 1980-1991. 18-6-09 2 Aim The aim of this paper is to describe and compare a number of methods used to measure the performance of products within a retail assortment. Based on this a new system will be suggested that should be able to reflect profitability better than existing measures. It should be possible to decentralize the authority to make decisions concerning gross margin percentages. Such a system must meet the condition that it is practical for non-food retailing. Expected results The research should result in an overview of the factors that influence the profitability of a store. The trade-offs between these factors should become obvious. For example, what the possibilities are to change margins without affecting overall store profitability. The following question should be addressed: How can the sales price of a product be set in such a fashion that it forms a reflection of the actual costs involved in selling that product? This thesis consists of three parts. The first part is an explanation of retail management accounting in general. The second part is a description of empirical cost research done at C&A Germany into the profitability of parts of the assortment. A theoretical revenue model (the Assortment Model) is also explained. Finally aspects of the two models combined, offer possibilities for calculating the profitability of individual items. This alternative pricing methodology is discussed as well as the ramifications for overall company planning. The first part is an explanation of retail management accounting. This part of the paper consists of the first two out of a total of five chapters. The first chapter is a general explanation of traditional retail accounting and merchandise management, focusing on systems which are used in bookkeeping. The second chapter explores modern systems such as Direct Product Profitability (DPP) and Activity Based Costing (ABC) which are used to calculate the productivity of different parts of a retailer‟s assortment. 18-6-09 3 The second part of this paper is a explanation of work done at C&A Germany to study the costs and profits associated with different classes of goods. The first paragraph is a description of C&A Germany with particular emphasis on the organizational structure and the way merchandise management was handled. This is followed by an explanation of the way the C&A research was done and how it overlaps with other methods discussed in Chapter 2. Finally this chapter will examine whether this system would lead to other merchandising decisions than more commonly used indices such as gross margin or movement indicators (Lusch et al., 1993). In Chapter 4, the Assortment Model developed by Kevin Corcoran of the Intercena International Economic Department. is explained. This model was meant to help analyze assortment profitability and support decentralized store level decision making on important aspects of retail such as price setting. The combination of the revenue focus of this model and the cost measurement in the C&A Germany research lead to a possible alternative pricing methodology, which is discussed in the third part. In chapter 5 the alternative pricing methodology is explained. The ramifications of changing the way costs are viewed are explored using practical examples of different articles. An important aspect of any system is the way in which it can be fit into all levels of planning and thinking within a company. That is the reason that this chapter also considers planning at total company level. The last chapter is the conclusion, which contains ideas for further research and action. 18-6-09 4 Chapter 1: Introduction to retail accounting The purpose of this chapter is to introduce the reader to some of the basic principles of food & non-food retail accounting. The first section considers retailing in general, the ways in which it differs from other industries and how it can be split into categories. This is followed by a discussion of the retail income statement and relevant terminology. A brief explanation of the retail method of inventory valuation, which is used by most large retailers, illustrates how retailers track their profitability from day to day. The role of the gross margin percentage in retailing in general and pricing in particular is considered. Finally, the use of stock turn measurement and its effect on gross margin levels in the retail industry is described. 1.1 Retailing and the value chain Retailing consists of the final activities and steps needed to place merchandise in the hands of endconsumers (Ghosh, 1994). Retailing, as an industry, has many characteristics that differentiate it from other forms of commercial activity. Three of these characteristics are of overriding importance (Zimmerman et al, 1990): 1. High volume of low-value transactions. 2. Broad exposure of merchandise to theft or damage by employees, customers and suppliers. 3. Constantly changing product offerings. These conditions, in turn, influence many aspects of business such as personnel, space utilization, pricing, organization structure, and technology. An important analytical tool that can be used to understand industries, is Porter‟s value chain model. The idea behind the value chain is that all functions have to work together to coordinate activities so that the organization can be effective. Porter differentiates between „supportive activities‟ and „primary activities‟. Each of the categories may be vital to competitive advantage depending on the industry. Porter asserts however that this value chain is applicable to all firms. All the categories of primary activities will be present to some degree and play some role in every firm (Porter, 1985). 18-6-09 5 Firm Infrastructure Support Activities Human Resource Management Technology Development Procurement Inbound Logistics Operations production Outbound Logistics Marketing & Sales Service Primary Activities Figure 1. The generic value chain (Porter,1985) Van der Kind (1996) has, however, described an alternative value chain that is perhaps a better fit for retailing. When this retail value chain is compared to the generic value chain, the most obvious difference is the place of buying (procurement). Procurement is considered by Porter to be a supportive function for industrial firms. He acknowledges that the total cost of purchased inputs as a percentage of firm value provides an important indicator of the strategic significance of procurement. In retailing the percentage is so high that Van der Kind considers procurement to be a primary activity in the retail value chain. This is especially true for fashion sensitive retailing. Firm Infrastructure Support Activities Human Resource Management Technology Development Property Buying (Procurement) /Marketing Inbound Logistics Outbound Logistics Store operations / sales Primary Activities Figure 2. The retail value chain (Van der Kind, 1996) Porter defines operations as activities associated with transforming inputs into the final product form, such as machining, packaging and assembly. Products which are bought by retail buyers are 18-6-09 6 in essence the products customers will be offered. The types of utility provided by retailers are time, place, and possession utility. Retailers are not involved in providing “form” utility which is done by manufacturers and processors (Lusch et al, 1993). This is the reason that the category operations has been shifted from production to sales in the retail value chain. 1.2 Categorization of retailers There is no single, accepted method of classifying retailers. Different authors use different dimensions and definitions of dimensions to describe different companies. At the same time retailers and their formulas are constantly evolving in response to the changing needs of customers. Classification therefore tends to become arbitrary in reality (Lusch et al.,1993). In general, retail can be split into two main categories, goods on one hand and services on the other. There are many outlets that offer both under one roof, think of financial services offered by many department stores. This paper, however, is solely concerned with the retailing of physical goods. Zimmerman et al. use three further dimensions to make a basic classification of goods retailing: structure, merchandise and size. Structure is either in-store retailing or shop-at-home businesses. Not all retailing takes place in store settings. The importance of catalogs and other types of nonstore retailing such as Internet shopping, TV promotions, door-to-door selling, telemarketing and vending machines increased in the 1980s and 1990s (Ghosh, 1994). Size is important because as a retailer grows beyond a certain size, business operations and controls will change. Classification by merchandise (Zimmerman et al ,1990): I. Food A. Grocery 1. Supermarket 2. Convenience Stores B. Dining 1. Fast food 2. Fine dining 3. Food service II. General Merchandise A. Department, Discount and Chain Stores B. Specialty stores 1. Apparel 2. Hard Goods 18-6-09 7 C&A Germany, which will be discussed thoroughly in Chapter 3 has been described as a formula that sells apparel in department-style stores (Management Horizons, 1992). The differences between apparel stores and grocery stores are also part of Chapter 3. Dining is basically a different industry and will not be considered in this paper. 1.3 Income statement This paragraph considers the retailer‟s income statement which forms the basis for planning and price setting. Five elements of the income statement summarize key dimensions of the firm‟s operations for the year. These are:(1) sales; (2) cost of goods sold; (3) gross margin; (4) selling, general and administrative (SGA) expenses (also called operating expenses); and (5) profit. It is important to first clearly define what each of these elements measure. A fictive income statement is shown below, followed by definitions of the key terms. Table 1 A retail income statement Gross sales Less:returns (1) Net sales Beginning inventory (at cost) Purchases during year Freight Inventory available for sale Less: ending inventory (2)Cost of goods sold (3)Gross margin Wages and salaries Rents and occupancy Selling expenses General expenses (4)Total SGA expenses (5)Net profit $272,000 __2,200 85,000 170,000 __1,000 $256,000 80,000 176,000 94,000 38,200 24,400 8,000 _6,000 76,600 17,400 $270,000 18-6-09 8 Sales: Gross sales reports the total amount of money obtained from customers through the sale of merchandise and services. The amount after accounting for returns and other adjustments is net sales. When a reference is made to a store‟s “sales”, it usually means net sales. Cost of Goods Sold (COGS) Cost of goods sold is usually the largest expense item for retailers. It tends be highest for supermarkets where it typically represents 80 to 85 percent of net sales. In department and apparel stores, the cost of goods is about 60 to 65 percent of revenues (Ghosh, 1994). The cost of goods sold is the total expense the retailer incurred to put together the merchandise assortment it sold the customers. Basically it is the amount paid to suppliers for merchandise sold to customers. It will also include any cost it incurred in transporting the merchandise to the distribution center and the cost of alterations. In Europe it is customary to include value added tax (VAT) in the COGS (Cox, Brittain, 1996). Gross Margin The difference between sales revenue and cost of goods sold is called gross margin or gross profit. This is an important measure because it indicates the amount of money that is available to cover operating expenses. Selling, General and Administrative (SGA) Expenses These expenses, also called operating expenses, include all expenses the retailer incurs except its cost of goods sold. Total SGA expenses can be divided into four major groups as shown in the table above. Wages and salaries represent the cost of paying all personnel working for the firm. Salaries as well as the cost of providing health, retirement and other benefits to sales personnel, clerical staff and managers are included in this category. Rent and occupancy covers the costs of renting the store site or the cost of the mortgage. Utility expenses and real estate taxes also fall in this category. Selling expenses are the direct costs of selling the merchandise. Advertising and promotion account for the major part of selling expenses. Some firms pay salespeople commissions based on the amount of sales they are able to generate. The amount of such commission is also included under selling expenses. General expenses include such items as depreciation and interest. 18-6-09 9 Profit: The final element of the income statement is net profit. This reports the amount the firm earned (or lost, if net profit is negative) during the period covered by the income statement. It is found by subtracting the costs of goods sold and total SGA expenses from net sales. Subtracting the amount of tax payable by the store from this figure yields the net profit after taxes. 1.4 The Retail Inventory Method To be able to draw up an income statement (and a balance sheet) retailers must be able to track and valuate the flow of goods through their operations. Ideally, records should be kept of the product cost price of each individual item bought, sold and in inventory at the end of the period. In section 1.1 Retailing and the value chain, some of the characteristics of retailing were discussed. Two characteristics mentioned were that retailing has to do with a high volume of low-value transactions and constantly changing product offerings. In the distant past small shops owners could keep track of their goods fairly easily and the selling prices were often the result of negotiation with the customer. With the arrival of mass merchandising, keeping track of the cost of goods of individual items became more complex. This was especially true for physical stock-taking (inventories). Large department stores such as the Bijenkorf can have as many as 200,000 SKU‟s (stock keeping units) at any one time (Van der Kind, 1996). Without the help of computers, there is a lot of work involved if every item has to be counted, its cost price (what the retailer paid for a product) and selling price noted. The cost price is furthermore considered to be competitive information, so it would have to be encoded on the label or linked to an item number which would entail a search in files for cost figures. The retail method was developed in the early 1900s in response to these difficulties. The retail method records the purchases at cost and sales at retail (selling price), recognizes price changes and then computes a collective estimated value for ending inventory at cost. The objectives of the retail inventory method are to determine interim inventory amounts and the cost of goods sold and to monitor inventory shortages. Under the retail method, the cost of goods sold for any given sale of an item is unknown. When an item is sold or when physical stock-taking is done, only the retail price is noted. The retail method is basically an averaging technique. While sales levels are recorded accurately, the value for ending inventory at cost is an estimate. 18-6-09 10 The retail method works by keeping records of transactions done in a certain category of goods. All purchases, sales and markdowns are recorded. The method estimates the historical cost value of inventory at the end of a trading period. Before the use of computers only the level of sales were recorded . The exact items sold were not registered. Stores did not know whether they had sold goods with a high or low margin or even no margin. The retail inventory method uses a cost multiplier which reflects the average relationship between the retail and the cost value of inventory. During a certain period the retail value of inventory and sales is tracked, subsequently at the end of the period the retail value is converted back to cost using the cost multiplier. This is illustrated in the next table. Table 2 Illustration of the retail method of inventory valuation (adapted from Zimmerman et al., 1990) Cost Cost Opening inventory Purchases Total merchandise handled Net sales Markdowns Shrinkage (theft or loss) Total deductions from retail value Ending inventory 5 684 (b) 12 000 47,37% 5 000 4 000 9 000 Retail price 9 000 10 000 19 000 6 000 900 100 7 000 47,37% (a) Multiplier (a) Cost multiplier = Total cost of merchandise handled = 9 000 = 47,37% Total retail of merchandise handled 19 000 (b)Ending inventory at cost=Ending inventory at retail xCost multiplier = 12 000 x 47,37%=5 684 The cost multiplier should not be confused with the realized gross margin. The complement of the cost multiplier is known as the cumulative markon (here 100-47,37=52,63%); it is higher than the gross margin. It is a percentage of the total merchandise handled at retail, while the gross margin percentage is expressed relative to net sales. The gross margin is sales minus cost of goods sold. 18-6-09 11 In this example gross margin would be: 6 000 (net sales) – (9 000 – 5 684) (cost of goods sold) = 2 684. The gross margin percentage is the gross margin divided by sales, 2 684/6 000 = 44,73 %. Had there been no markdowns taken or shrinkage in the period than the cumulative markon would be equal to the gross margin percentage. There are different ways of working with the retail method. The cost multiplier, for example can be calculated on the basis of the opening inventory without taking purchases into account (Lusch et al.,1993) or at the other extreme, based solely on the purchases in the period (Zimmerman et al,1990). In the calculation used above both opening inventory and purchases are taken into account. Understanding the arithmetic and methodology of the retail method is not essential for this thesis. It is, however, important to understand that most retailers work with dollar inventories and gross margins which are the averages of a series of items. Some retail consultants, given the sophisticated inventory systems available today, are recommending a switch back to the cost method (Lusch et al., 1993). Software suppliers are offering merchandise programs which do track cost of inventory, retail prices and sales at item level and these are being used by supermarkets, mass merchandisers as well as some apparel stores. Zimmerman saw the possibility of the retail inventory method being replaced or augmented by systems like direct product profitability (DPP). At present, though planning and pricing in retailing in general is still strongly based on storewide average gross margins and selling, general and administrative (SGA) expenses expressed as a percentage of net sales. This is illustrated in the next section. 1.4 The gross margin and price setting Of the available profit measures or operating ratios in retailing, gross margin is the most important (Zimmerman et al., 1990) and most widely used in retailing to assess merchandising performance (Lusch et al., 1993). There are two reasons for this. First it is the most accurate figure to work with, since both sales and cost of merchandise can be measured by merchandise line with minimal error. Second, many industry trade associations regularly report data on gross margins by merchandise line, making it possible for the retailer to compare its performance to the experience of others (Lusch et al., 1993). It is also used by outsiders to judge the performance of retailers. 18-6-09 12 Van der Kind (1996) maintains that the gross margin percentage is an essential part of understanding all retail operations concerned with the sale of goods. He claims that the key to retailing is attaining a profit through maintaining sales at a certain gross margin percentage. In order to create sales, expenses have to be made. He starts out from the following formula when analyzing retail operations: Profit = sales x gross margin percentage – SGA expenses This formula results in the same answer as one used by Ghosh. Profit = sales – cost of goods sold –SGA expenses Ghosh uses absolute numbers, not a single gross margin percentage for the whole retail operation. Van der Kind‟s formula seems to be based on thinking influenced by the use of the retail inventory method, in which the cost of goods sold is not tracked directly, but is the result of calculations based on an average cost multiplier. The use of the retail method is, however, not the only reason companies use company wide margins when planning. Nooteboom (1982) believes that to a large extent “customary margins” are employed simply because it is virtually impossible to attribute the costs of operating a store to individual items in the assortment, and to assess price elasticities between goods as a basis for optimal pricing by an equation of marginal costs and marginal revenues. Van der Kind (1996) uses the same arguments, concluding that in practice retailers still use company wide margins when setting prices. The use of customary margins for price setting is known as cost orientated pricing or the "cost-plusmodel". In theory no account of competitors or demand is made. Two other types of pricing that can be used in retailing are demand orientated pricing and competition orientated pricing: Under demand orientated pricing the price is set at a level deemed commercial independent of cost. Retailers using competition orientated pricing set prices at or under their most aggressive competitors level (with a higher volume or lower overhead it is possible to offer lower prices)(Cox, Brittain, 1996). Traditionally the first step in pricing is to set a general margin to be achieved within a store or department. This goal is established by determining the amount of operating (SGA) expenses necessary to achieve the forecasted sales volume, as well as the profits desired from the specific 18-6-09 13 operation (Stern and El-Ansary, 1992). The ways retailers regard prices on the one hand and customers on the other is illustrated in figure 3. Level of perceived utility Price Value or consumer surplus Retailer‟s profit Amount allocated to cover SGA expenses Cost of goods sold (COGS) Gross margin Cost of goods Figure 3. Utility and price (adapted from Ghosh, 1994) Figure 3 shows how retail prices relate to costs, on the one hand, and consumer utility on the other. The utility represents the usefulness for the customer buying an item. If the utility doesn‟t exceed the price, the article will not be sold and the article will have to be marked down or dropped from the assortment. The difference between the retail price of a merchandise item and its cost to the retailer is the gross margin. Part of the margin covers the selling, general and administrative (SGA) expenses, and the remainder is the retailers profit. Retail pricing decisions rely heavily on the markup concept. The process starts with the income statement for the entire company. Based on the expectations and goals for the coming year, companies set a general sales target as well as a company wide gross margin target. These are based on estimations of the SGA expenses the company foresees and it‟s profit target. 18-6-09 14 The calculation of the margin target: (Ghosh,1994) Margin = SGA expenses + profit Margin percentage = (SGA expenses + profit)/sales x 100 The goals for the store are divided and allocated to individual departments within the store and individual merchandise classifications within departments to guide them in setting prices. There are always exceptions to the rule that the margin is uniform throughout the merchandise mix. The best known example of these items are the so called loss leaders or what Van der Kind refers to as “Key Items”, which are used to attract customers to a store. Market research indicates that consumers are more sensitive to price changes of some products than of others. These are items offered at low prices, close to cost, to create traffic (Van der Kind, 1996). A quote from the founder of Wal-Mart illustrates this thinking, ”Health and beauty aids were priced to give away. Everything else in the store was priced low too, but it had a 30 percent margin ”(Walton, 1992). Margins for different categories will vary depending on demand and competitive conditions. Although the price set for a particular merchandise item may differ from the overall goals, aggregate markups must relate pricing decisions to overall profit goals (Ghosh, 1994). This is done in such a way that a storewide markup percentage is obtained. To achieve this, some categories may be priced with considerably higher markups and others with substantially lower markups than the storewide average (Lusch et al., 1993). There is a difference, however, between the initial price that retailers request for their goods and the amount they eventually receive. For fashionable articles especially, many items have to have their price reduced before they are sold. This is the reason for the difference between the initial markup, the percentage which is used to set prices and the gross margin (also known as the maintained markup). Retailers do not casually arrive at an initial markup percentage: it must be carefully planned (Lusch et al., 1993). Within a store there can be categories of goods that have little discrepancy between initial markup and gross margin, while other categories may have a difference greater than 20 %. At C&A, kitchen aprons are an example of the former and fashion accessories of the latter. Initial markups must 18-6-09 15 provide for markdowns, shortages, employee discounts and alteration expenses (together referred to as total reductions), which reduce net revenue (Lusch et al., 1993; Stern and El-Ansary, 1992). Markup goal = Expenses + profit + total reductions Net sales +reductions Consider the following example: jewelry bought for $5, SGA expenses are $4, profit target $1, reductions of $2 are expected. The first selling price is $12 ($5+$4+$1+$2), the goods are actually sold for $10 ($12-$2). Initial markup = $4 (Expenses) + $1 (profit)+ $2 (reductions) = $7/$12 = 58% $10 (Net sales) + $2 (reductions) Gross margin percentage = 4 (Expenses) + 1 (Profit) = 5/10 = 50% 10 (Net sales) In summary, the initial and gross margin percentages contribute important information for price determination. Based on expected sales, expenses and profit targets, they indicate the margin the store must generate to meet its overall financial goals and serve as guidelines for determining the prices of individual items. Retailers do not, however, apply them to determine prices of all items in a store. Markups for individual items will vary depending on demand, competitive conditions, consumer price sensitivity and the marketing environment. 1.5 Stock turnover and Volume versus Margin Next to gross margins, stock turnover is considered to be one of the key elements in retailing. Stock turnover, which is also referred to as merchandise turnover or inventory turnover, is a concept which evolved over a century ago. Before the expansion of railway lines and telegraph systems, retail was a relatively slow and small scale process. Store managers would buy their goods once a season and travel back with them to their establishments. At the same time traveling salesmen would go door to door with their wares. The telegraph, post and trains made it possible for stores to sell goods which could be reordered and delivered during the season. Mail order catalogues such as those produced by Sears in the United States were also made possible. Small 18-6-09 16 independent stores were furnished with goods by increasingly large wholesalers such as Marshall Field. Wholesalers bought goods from producers in bulk, stored and them distributed them from large warehouses often situated near rail terminals to small independent shops. The idea of stock turn was conceived by these wholesalers in the 1890‟s (Ortega, 1998). Stock turnover described how many times the stock on hand was sold and replaced within a certain time period, usually a year. Thus, a stock turnover of 12 indicates that, on average, the retailer turns its inventory 12 times each year or once a month. The different ways of measuring this which will be discussed in Chapter 2. The higher the stock turnover with a given number of workers and the same facilities, the lower the cost per unit sold. It was a way to measure efficiency. Department stores such as Macy‟s in the U.S. and Marks & Spencer in the UK adopted this strategy of focusing on stock turnover as well. It was around this period in the 1880‟s that C&A as well started to deviate from certain traditions in the retail trade in Friesland: “(C&A) undersold its competitors and the higher turnovers could offset the lower profit margins” (Surrey Beheermaatschappij bv, 1973). As was stated in section 1.2, Categorization of retailers, there are many ways of classifying different retail formulas. One method, not yet discussed, is segmentation between on the one hand, high-margin, low-turnover formats and on the other hand, low-margin, high-turnover formats. Both formats continue to exist, but in the twentieth century the spotlight focused on the revolutionary volumes flowing out of the latter style of operation (Stern and El-Ansary, 1992). Essentially, the low-margin, high turnover model is orientated toward generating maximal operating efficiency and passing on the savings generated to the customer. According to Stern and El-Ansary, however, many of the savings “passed on to the customer” must be seen as a shifting of cost instead of an elimination of it. Customers are expected to do many of the things retailers once did for them, for example carrying groceries home, or finding and selecting merchandise. On the other hand it is possible to argue that sustained high sales volume can be used to cover the expense of service normally expected at higher margin establishments. The idea is stated simply at Wal-Mart, “Make your profits by selling more goods, instead of selling goods for more.” In the long term Wal-Mart would like to have sales of $300 billion, about six times total retail spending in the Netherlands in 1994 (Ortega, 1998; Van der Kind, 1996). 18-6-09 17 The first Wal-Mart store opened its doors in Arkansas in 1962 (Walton, 1992) within months of the introduction of Kmart (for many years the biggest discounter in the U.S.) by the Kresge company. In the same year, 1962, Carrefour invented the hypermarchė in France (Ortega, 1998) and Theo Albrecht opened the first “real” Aldi store in Dortmund Germany (Brandes, 1998). These stores and others had a great effect on the way retailing is done, especially on the steady reduction of the level of gross margins. An independent grocery store in the 1950s would have operated off a 60-70% gross margin, based on buying from a wholesaler already taking a 30% or greater markup on the manufacturer's selling price. Mom-and-Pop Independent USA/UK/France in: 1950s Store sales (1996 $) 100 000 Retail gross margin 50-60% Wholesaler gross margin 25% Retail price multiple 2.7-3.3x Table 3 The trend in gross margins from (Wileman&Jary, 1997) Small scale supermarket 1970s 2 000 000 25-35% 10% 1.5-1.7x Large scale supermarket/ warehouse club 1990s 100 000 000 10-20% N/A 1.1-1.3x As retail sales moved from small independent stores supplied by wholesalers to large chains of stores who bought directly from suppliers, margins dropped. Increased competition and the high sales growth targets kept pushing the amount retailers could charge for their services down. The percentage of gross margin in the discounting industry dropped steadily from around 35 percent in the early sixties to only 22 percent at the beginning of the decade. Gargantuan, selfservice warehouse clubs such as Priceclub, Costco and Makro, operate in the U.S. at gross margins of between 5-7 percent of sales with an additional income provided by a membership fee of around $25 a year (Walton, 1992). The following chapter, Cost calculation and productivity measurements, discusses ways of evaluating retail operations and the products sold in the merchandise mix, which are more specific than the traditional use of company wide margins and stock turn. 18-6-09 18 Chapter 2 Cost calculation and productivity measurements This chapter discusses two different systems used to evaluate merchandise performance, the Trinity model and Direct Product Profitability. The Trinity model (or strategic resource management model) is based on work done by Lusch, which combines three ways of looking at store performance. It is known as the Trinity model because it provides a framework for strategically managing three key retail resources. The first sections of this chapter focus on the productivity of these resources: 1. Inventory (the bulk of current assets);Gross Margin Return On Inventory Investment (GMROI) 2. Selling space (fixed assets); 3. Labor (human resources); Gross Margin Return On Space (GMROS) Gross Margin Return On Labor (GMROL) These three elements are then combined into the Trinity model in the fourth section. The second system discussed in this chapter resembles Activity Based Costing. This is the direct product profitability (DPP) calculation, which became practical with the advent of computer spreadsheets. 2.1 Gross margin return on inventory investment Retailers use a number of different measures to monitor merchandise performance, two of the most common of which are gross margin and stock turnover. These measures were discussed in Chapter 1. Gross margin return on invested inventory (or GMROI) combines both measures into a single figure. There are different ways of calculating GMROI, that all deliver the same result. Starting from the gross margin percentage, it can be calculated as follows: GMROI = Gross margin percentage x Sales to stock ratio The sales to stock ratio is an expression of stock turnover : Sales to stock = Net sales/Cost of average inventory Sales to stock is higher than stock turnover in units as discussed in Chapter 1. Net sales is expressed in retail (selling price) dollars and cost of average inventory is the value of the inventory at the price the retailer paid for it, which should be lower than the retail price. 18-6-09 19 The GMROI rate combines the effect of profit margin and stock turnover in one performance measurement indicator and thus facilitates equitable comparisons of the profit contributions from different merchandise investments (Zimmerman et al, 1990). GMROI can be equal for stores using different strategies. The table below could be a comparison between a boutique with high margins and discounter with lower margins and a higher stock turnover. Table 5 GMROI result for high margin, low stock turn and low margin, high stock turn formats Boutique Discounter Gross margin percentage 45% 25% X X Sales to stock 4 7,2 = = GMROI 1,8 1,8 If the two stores in the table have comparable operating (SGA) expenses and the same amount invested in inventory, they will be equally successful. GMROI reflects the amount of money the retailer retains annually to pay expenses and yield a profit for each dollar invested in inventory. It measures the productivity of assets invested in merchandise inventory. Thus, a higher GMROI indicates a more productive item than a lower GMROI. Retailers typically set GMROI targets for the entire store and for different merchandise classifications. Many retailers evaluate the profitability of each merchandise item separately based on these criteria in order to identify poor performers and eliminate them from the merchandise mix (Ghosh, 1994). In addition to using GMROI in order to make decisions whether to add or drop certain articles from the merchandise mix, it is possible to improve inventory productivity on existing products. Zimmerman suggests two possible strategies to increase GMROI: reducing inventory investment and changing the selling price. Average inventory investment can be reduced by cutting down the stocks in the pipeline such as in the reserves, distribution centers etc. Good logistics and communications systems are essential ingredients in controlling inventory investments. Many observers believe that a key reason for WalMart‟s success is their state-of-the-art distribution and logistics system (Ghosh, 1994). Changing the selling price has a direct effect on the gross margin. Depending on the price elasticity of the product and expected competitor response a decision can be taken to either raise or lower the selling price. 18-6-09 20 Increasing the selling price to increase the gross margin can lead to a higher GMROI if demand does not fall too sharply. This is the case for articles with a low price elasticity. A decrease in selling price to stimulate turnover is logical if the article(s) sold have a high price elasticity and competitors are not expected to react to a price change. Stefan Person, CEO of Hennes & Mauritz, described the effects of this strategy for his company: "In 1996 we decided to reduce our prices, which made the difference between our prices and those of our competitors even greater. Lowering our prices boosted sales by 28% in comparable currencies. Despite lower prices earnings rose 44% over last year. Net margin improved thanks to a much higher level of sales without a corresponding increase in costs," (Hennes & Mauritz, Annual report 1996). 2.2 Gross margin return on space Many retailers do assortment planning using only gross margin on invested inventory. Thus, a stock keeping unit (SKU) or group of SKUs with a high GMROI receives more merchandising emphasis. However, this can be a questionable emphasis at times. If a store has very little money invested in inventory, the gross margins earned might be substantial compared to that investment, but retailers have to cover their Selling, General and Administrative (SGA) expenses in order to make a profit as discussed in Chapter 1. To illustrate this consider two fictional stores selling trousers situated next to one another. Their SGA expenses are around $50, 000,- and they are identical except for the product they sell. Store selling: Polyester trousers ($19) Woolen trousers ($49) Gross margin 50% 36% X X Sales to stock 10 7 = = GMROI 5 2,5 One store sells polyester trousers at a high margin and with a high stock turnover. The other, selling wool trousers, is less successful in managing these aspects of retailing. The selling prices are polyester trousers $19,- for the polyester trousers and $49,- for the woolen trousers. Both stores have an average stock of 1,000 pairs. The value of inventory at cost in the polyester store is product cost price per pair multiplied by the number of pairs in stock $19,- x (1-0,50) x 1000 = $9 500. The value of the woolen trousers at cost is $49,- x (1-0,36) x 1 000 = 31 360,-. 18-6-09 21 Gross margin dollars = GMROI x Cost of average inventory Gross margin dollars for polyester = 5 x 9 500,-= $ 47 500,Gross margin dollars for wool = 2,5 x 31 600,- = $ 79 000,If the SGA expenses are approximately $50 000,- the store selling polyester pants will be making a loss, while the store next to it with the lower GMROI is profitable. The store selling polyester trousers uses it stock more productively but the value of the polyester trousers is so low that the return on it is not sufficient to cover the store's expenses. The store with the woolen pants is earning more money within a similar store because it is using it's space more effectively. Many authors believe that the key resource with which a retailer should be concerned in planning assortments is space, not dollars of investment (Lusch et al., 1993; Drucker, 1980). This was also true for C&A; "Meters are the Αlpha and Omega of our business", as will be discussed in Chapters 3 and 4. When presented in a similar way, the woolen trousers have a higher investment per unit of space occupied than polyester trousers. Investment per unit of space is known as inventory intensity. It is the cost of inventory per square meter . The SKUs that contribute most to effectiveness of an assortment are those that have a high inventory intensity and also have a high GMROI (Lusch et al., 1993). The three factors (gross margin, sales-to-stock ratio, and inventory intensity) when multiplied together yield gross margin return on space (GMROS). GMROS = gross margin percentage x sales to stock ratio x inventory intensity = gross margin dollars/sales x sales/cost of inventory x cost of inventory/sq. meters selling space and thus: GMROS = gross margin dollars / square meters of selling space The following example illustrates the implications a shift in emphasis from GMROI to GMROS can have. Supermarkets are faced with a practical dilemma when choosing whether to stock expensive well known producer branded products such as those from Proctor & Gamble or their own private store brands. Those who concentrate on cheap store brands usually focus on margin 18-6-09 22 enhancement because their gross margin percentages tend to be higher than on strong producer brands. On the other hand more expensive products entail a higher inventory intensity. “As producers are at pains to point out, cheap store brands may deliver higher percentage gross margins, but rates-of-sale per line and per square foot are usually much lower than for producer brand alternatives, so that the critical measure of $ gross margin per square foot of selling space is lower than on the low gross margin producer power brands.” (Wileman & Jary, 1997) The different strategies of Kmart and Wal-Mart can be interpreted as a difference in focus between inventory productivity (GMROI) as opposed to space productivity. “Kmart turned more and more to private-label goods, items made for Kmart and sold as a Kmart house brand. Private-label goods can carry a fatter profit margin, because retailers typically pay the maker much less for them than for similar name brands. The problem though was that many of Kmarts private labels were not very good. A lot of the clothing was shoddy, even ugly. By the mid-70‟s Kmart was being nicknamed the Polyester Palace…” (Ortega, 1998). At Wal-Mart, on the other hand, every executive has to choose and sponsor an item called the VPI (or Volume Producing Item). Wal-Mart is focussed on increasing sales on the existing sales area. Suppliers design products for Wal-Mart, such as snow sleds, so they can be stacked on top of one another to increase inventory intensity. “… you have to be merchandise driven. Otherwise you become like everybody else... it is at the heart of what creates our extraordinary high sales per square foot, which enable us to dominate our competition.” David Glass, CEO (Walton, 1992). In 1991, Wal-Mart sales per square foot of retail space were $227. K-mart, in contrast, sold only $139 per square foot annually. (Lusch et al., 1993) 2.3 Gross margin return on labor Labor is one of the three main resources a retailer utilizes in its operations (besides inventory and space). The gross margin return on labor (GMROL) is the number of gross margin dollars earned per full-time-equivalent employee. GMROL can be obtained by multiplying the gross margin percentage by the labor productivity ratio (net sales divided by full-time-equivalent employees). In practice few retailers use the gross margin return on labor in the same way as GMROI and GMROS for comparing merchandise lines or individual SKU's. This is due the fact that individual employees may be involved in selling a multitude of different products during the course of one 18-6-09 23 day. It is therefore very difficult to allocate labor costs to individual products. When this is done it is usually in the context of a comprehensive costing system such as Direct Product Profitability which is discussed in the following paragraph. For certain retailers it is possible to allocate labor to products fairly accurately. The Aldi supermarkets, for example, use the following calculation to estimate labor costs for a single SKU. Labor is allocated according to the amount of shelf space an article uses and the number of times that shelf space has to be restocked in a certain time period (Brandes, 1998). L=l x STy x M2y 2 . STy x M y+ UHs-y x M2s-y L= Laborcosts for SKU y l= Total labor costs for the store excluding service departments STy= monthly Stock Turn for SKU y STs-y= monthly Stock Turn for the total self-service assortiment excluding SKU y M2 y = Shelf length allocated to y M2s-y=Total amount of self-service shelfspace in store excluding y This system works for Aldi because Aldi stores have certain characteristics which minimize the variations when handling different products. The formula's merchandise mix includes very few fresh products and the number of articles is limited to 600. Products are not handled individually; instead standard sized cardboard boxes are used throughout the store. In contrast conventional supermarkets have certain shelves with many articles, such as those containing candy, which require much more attention per square meter than other shelves stocked with bulk goods such as toilet paper. In general gross margin return on labor is used to compare stores or chains of stores with one another. A high GMROL can indicate efficient operations and motivated staff. It is also possible to make trade-offs between inventory or space productivity on one hand and labor productivity on the other. Inventory productivity can be increased by having a minimum amount of stock in the store, but that entails frequent stock counts and replenishment if stock outs are not to occur. On the other hand labor productivity can be increased by moving large amounts of a single product into the store at one time. Warehouse clubs sell directly off of pallets, following the principle "stack it high, sell it low", this increases labor productivity but leads to lower inventory productivity (GMROI) than might otherwise be possible (Van der Kind, 1996). 18-6-09 24 There is also a relationship between space productivity and labor productivity. If a store of a certain size increases the number of employees serving the customers, an increase in sales and thus space productivity can be expected. Increasing the number of employees however in general leads to lower sales per employee. The link between the space productivity and labor productivity is what Lusch describes as employee intensity, but is more accurately named the self-service ratio by Van der Kind. Self service ratio = Square meters of selling space/Full-Time-Equivalent Employees As the number of square meters in relation to the number of employees increases, customers will find fewer and fewer employees engaged in active selling. The self service ratio (M2/FTE) multiplied by the space productivity ($/M2 ) results in labor productivity ($/M2 x M2/FTE = $/FTE). A rise in the self-service ratio without a decline in space productivity will lead to an increase in labor productivity. 18-6-09 25 2.4 The Trinity model Using the Trinity model shown below it is possible to see the links between the three productivity measures discussed in the previous paragraphs. Net Sales Inventory Investment Gross Margin Dollars Inventory Investment GMROI Inventory Investment Sq. Meters of Space Gross Margin Percentage Net sales Sq. Meters of Space Inventory Intensity Gross Margin Dollars Sq. Meters of Space GMROS Sq. Meters of Space Full-Time- Equivalent Employees Net Sales Full-Time-Equivalent Employees (adapted from Lusch et al., 1993) Self-service ratio Gross Margin Dollars Full-Time-Equivalent Employees GMROL The Trinity model does not provide a direct estimate of profitability. Without extra information it is impossible to see whether a store or part of it is profitable. Information on the cost of capital, rent and occupancy as well as salaries is not included in the model. What it does do, is highlight the causes of negative trends and enable a comparison between different stores within a chain or between competing formulas. 2.5 Direct product profitability This paragraph discusses what Direct Product Profitability (DPP) is and what the difference is between DPP and the Trinity model. The evolution of the system is then described as well as how 18-6-09 26 it is used. Finally the difficulties of using DPP in the (apparel, fashion) non-food sector are described. Direct Product Profitability (DPP) can be defined as those systems which calculate (variable) direct costs which are connected to a certain product in the assortment. These costs are subtracted from the gross margin generated by the product. DPP = gross margin -direct product costs. A DPP analysis is not used to compare on store to another store as the Trinity model is often used, but to compare different SKU's to one another. Both systems are based on the principle of optimizing the use of scarce resources. The Trinity (strategic retail resource) model looks at the relationships between sales, margins and resources and how they affect one another. DPP reflects inter-product differences in sales, margins and costs associated with storing transporting, shelving and laborintensive merchandising activities (such as pricing individual items)(Borin and Farris, 1990). The output is a single number for each article. The DPP system works as a sort of black box: you can see what the profit is but not why it is what it is. The idea of allocating actual selling, general and administrative (SGA) expenses to individual products in retailing is relatively new. There are two reasons for this. It was considered impractical for a retailer to do cost accounting, because of the high volume of various low value transactions involved. The costs related to tracking and allocating expenses didn't weigh up to the expected benefits. Allocating expenses is also often deemed unnecessary for goods which are already in their final product form. Retailers are usually not involved in providing "form" utility through activities such as machining, assembly and packaging which is done by manufacturers. The types of utilities provided by retailers are time, place, and possession utility, as discussed in Chapter 1. The majority of goods are purchased ready for sale and expense measurement is related to cost of purchase (COGS). An average cost figure is charged against the gross margin. Some departments in stores such as supermarkets transform materials into finished goods. Only for these operations such as inhouse bakeries, was a system for accounting for the additional costs of transformation processes considered necessary (Arnold & Turley, 1996). 18-6-09 27 It is interesting to note that the some of the first efforts to measure actual products cost of finished goods within retailing were sponsored by manufacturers, who were themselves used to working with management accounting systems. A study of profits and expenses of the grocery business was conducted in 1963 by the consulting firm McKinsey & Co. for General Foods Corporation, America's largest producer of processed foods. The study was based on a transactional analysis of business costs. The Mckinsey study showed that actual costs depend on the transactions each commodity needs, and that the share of the expense burden of different commodities varies greatly (Drucker, 1964). The data and number of calculations needed to perform the comparisons in combination with the scarcity of computers at the time made the system unfeasible for everyday use. Transaction based costing evolved and is now known as Activity Based Costing (probably because the abbreviation ABC is more appealing than TBC). Developments in retailing led to a greater need for the accurate allocation of expenses. New store formats with mixed inventory assortments and heavily promotional or competitive pricing such as modern hyper- and supermarkets have been most sharply confronted with the limitations of the retail inventory method (Lusch et al., 1993). The McKinsey study had demonstrated that commodities within a mixed assortment place unequal burdens on a store's resources. Promotional and competitive pricing make it difficult for stores to adhere to a single cost multiplier (and thus gross margin) when setting prices. Products in a hypermarket can take up different amounts of space, have a wide range of selling prices, varying stock turnover speeds and totally disparate handling needs. Traditional measures of merchandise productivity aren't sufficient for stores selling items as diverse as sandwiches and furniture under the same roof. In the late 1970's and early 1980's advances in hardware and software led to the availability of inexpensive and accessible computing power so that large amounts of data could be processed quickly and accurately. At the same time retailers started using optical scanning equipment to collect real time information on stock levels and revenue at the point of sales (POS). Cash registers fitted with scanning equipment to read barcodes developed into data registration units capable of recording the daily sales of SKU's at store level. 18-6-09 28 The combination of data collection and the ability to process it, was the impetus for McKinsey (now together with Proctor & Gamble) to re-introduce the concept of direct costing in the supermarket (Osté-Ruizendaal, 1994). They built a model for supermarkets using the practical PC application Lotus 123. The motivation for Proctor & Gamble was to convince supermarkets to look further than gross margin percentages when making merchandising decisions because private labels seem quite profitable on that scale (as stated in Section 2,2). Other manufactures and suppliers followed suit and introduced their own models, which led to some confusion as to which expenses to include and how to calculate them. In the United States the Food Marketing Institute introduced the "unified" DPP Model in 1986. It developed by a working group of producers, distributors and supermarkets and quickly became the standard. 2.5.1 Calculating DPP Calculating Direct Product Profitability closely resembles Activity Based Costing (Janse-de Jonge, de With, 1999). Direct costs are allocated to individual products based on the activities carried out in the process of selling them to the customer. In general between 60% to 80% of expenses are accounted for, overhead expenses (such as head office) are not allocated. The sum of the assortment's Direct Product Profit should be sufficient to cover these costs and the company wide profit target. Direct Product Profit is calculated by subtracting a product's Direct Product Costs from its adjusted gross margin. The gross margin (sales minus cost of goods sold) is adjusted for other revenues such as discounts and allowances which are customary in the supermarket branch. The Direct Product Costs are calculated as follows. The model consists of activities or components which each have standard costs. The standard costs are the costs incurred in performing a certain activity. The costs connected to a certain SKU are calculated by relating the standard costs to product inputs such as weight, volume, sales rate, delivery method and pallet dimensions. An example of a cost component is the opening of cardboard shipping boxes containing pots of jam in a store. If the standard cost is $ 0,05 per box and there are 12 pots in a box (product input), the direct product cost will be $ 0,05/12 units = $ 0,004 per unit. The costs of all the components needed in selling the product added to the inventory investment expenses equals the Direct Product Costs of the product. 18-6-09 29 Direct Product Profit Computation (Borin and Farris, 1990) Sales _ Cost of goods sold Gross margin + Other revenues Warehouse Direct Costs Adjusted gross margin _ Direct Product Profitability + Transportation Direct Costs Direct Product Costs + Store Direct Costs The above figure splits the cost components into the areas where they are incurred. They can also be classified by type. The three main types of expense in the DPP model are: handling costs, space costs and inventory. Handling includes order picking in the distribution center, filling shelves in the store and cash desk processing. Space expenses are incurred when products take up space in the distribution center, store storage area or on the shelves. Inventory investment expense depends on the amount of money invested in the product. It is important to note the difference between sales to stock which is used for GMROI (gross margin return on investment) calculations and the inventory investment expense calculated for DPP. In DPP the payment terms of when the goods are actually paid for, are taken into consideration. Products that are in the store for a long time but are only paid for after a few weeks may have lower inventory investment expenses than fast moving products which are paid for on delivery. (The issue of payment term differences is irrelevant for C&A which uses standard terms for all its suppliers). 18-6-09 30 One of the main difficulties in using DPP is deciding on the activities to be included. Collecting the necessary data to set up standard costs for the cost components is also expensive. Calculating standard costs can involve time and motion studies and around 80 different inputs. Many retailers consider information on the efficiency of their operations confidential and are not willing to share it with others. Suppliers of modern DPP models pre-load them with industry labor and cost standards to make initial implementation easier. An example of this is the Category Profit (& Cost) Model, "The Practical Activity-based Costing Tool for Category Management" developed the Willard Bishop Company (which wrote the manuals for the "unified" DPP model) and Proctor & Gamble (WBC,1997). The system still requires a great deal of maintenance especially for changes in product assortment. In the supermarket branch assortment changes take place for about a fourth or a third of the assortment every year (Van der Kind, 1996). 2.5.2 DPP in practice DPP can be used to support decision making in merchandising and distribution. Merchandising decisions depending on DPP can include whether to add or drop an article from the assortment, pricing and space allocation and finally decision on which items to promote. Articles with a below average DPP should get less space allocated to them and products with a higher DPP should receive more space or a better space in the store for example at eye level (Borin and Farris, 1990). DPP can also be used to help decide between alternative methods of distribution and evaluate possible new scenario's through simulation. Supermarkets receive goods from their own distribution centers as well as directly from suppliers (direct store delivery) DPP can quantify the (financial) effects of changing delivery routes for both vendor and retailer (Brenninkmeijer, 1996). Direct product profitability has clearly had an influence on the supermarket industry. DPP is one of the most well known and used abbreviations in the industry. In conversations between manufacturers and supermarkets phrases like, "it's good for your DPP" are frequently heard (Van Abswoude, 1997). Item level activity is important since it is at this level that stores and their customers actually buy, but it was not usually the level of control in the past. Advances in technology are changing merchandising and operating techniques. These developments are just beginning to be felt in the accounting areas, but it is reasonable to expect that the Retail Inventory 18-6-09 31 Method may be replaced or augmented with more specific, cost accounting like techniques such as DPP (Zimmerman et al, 1990). The result might be a step away from using cost of goods sold in conjunction with an average gross margin to determine selling prices. On the other hand relating prices to actual expenses is considered unrealistic by other authors. Van der Kind maintains that the number of articles is too high, "Even a specialist retailer has an assortment which is much larger than that of the average industrial producer. To set up a separate cost calculation for each individual product is very prone to the judgement of those putting it together and practically impossible." In his opinion companies use DPP as a sort of reversed cost calculation, but looking forward they usually calculate prices with the traditional cost multiplier. Setting prices based on costs calculated using DPP or by using a cost multiplier are both examples of cost orientated pricing, as discussed in Chapter 1. Some stores are exponents of competition orientated pricing. They maintain that retailers cannot simply ignore market prices. Otherwise they risk pricing themselves out of the market when using incorrect calculations or an ineffective cost multiplier. Knowing your selling expenses per item can support competitively advantageous pricing decisions. "Aldi sells a Margaux red wine for 17,98 DM, competitors using a traditional cost multiplier (based on an average gross margin) sell this wine for at least 35,- DM" (Brandes, 1998). At Aldi the absolute dollar markup on wine is comparable to that of water. Expensive items have a higher inventory intensity than less expensive items. A single bottle of wine represents many times the value of a single bottle of water, so the gross margin percentage for wine can be set significantly lower than for water, while achieving the same DPP. 2.5.3 Application of DPP in the non-food sector Direct Product Profitability and related models (connected with Category Management and Efficient Consumer Response) have been used almost exclusively by hyper- and supermarkets. Non-food retailers in general have hardly taken part in these developments. A survey in the Netherlands of the companies Amici, Bijenkorf, HEMA and V&D reported that none of these companies were involved in DPP (or similar models) to calculate cost information at SKU level (Janse de Jong, De With, 1999). 18-6-09 32 There are a number of reasons DPP is considered difficult in non-food retail (various authors):    Costs are perceived to be too high in relation to benefits, margin structure is different Assortments are constantly changing and larger than in supermarkets Lack of fixed (shelf)space One of the reasons Vroom & Dreesman doesn't consider DPP to be worthwhile is because department stores utilize higher cost multipliers than most supermarkets. Getting the right styles and colors into the store is considered to be the most important aspect on which to focus. The following table illustrates how the margin structures of retail formats can differ. Table 6 Main components of retail margin structure: Fashion example: Sales @ full list price Sales Full price gross margin Markdown/clearance Realized gross margin Store-level costs (labor, property, other) 100.0 65.0 -15.0 50.0 -30.0 20.0 -10.0 10.0 Supermarket example: Sales @ net realized price 100.0 N/A N/A 25.0 -15.0 10.0 -5.0 5.0 Store contribution Central costs (logistics, marketing, head office) Operating profit Wileman&Jary, 1997 Product risk (having the wrong colors or styles) is the primary driver for the differences in initial markup (full price gross margin). Most fashion specialists clear a high proportion of stock at deep markdown prices, and their initial, full price margin structure of 60% or more reflects this. While other retail sectors have to manage some level of product risk (for example, perishable fresh product for supermarkets), it is not as severe and does not require the same gross margin “cushion”. Store-level scale and sales volume leads to the difference in realized gross margin. Store-level costs (rent, staff, store overhead) are a high percentage of sales for many non-food retailers operating out of small outlets with low absolute sales volume; supermarkets work off much lower store-level cost/sales ratios (Wileman & Jary, 1997). Central costs are higher because assortments have to be changed regularly and many fashion retailers are involved in activities such 18-6-09 33 as designing a collection. At C&A 150 buyers were responsible for sales of 2,7 billion DM in 1972, while at Aldi a single central buyer is currently accountable for between 2-5 billion DM of sales (Eglau, 1972; Brandes, 1998). The second reason DPP is not considered applicable to the non-food sector is that assortments are constantly changing and larger than most supermarkets. Clothing and many other products can be very seasonal, the assortment which is sold in the winter is generally very different from that sold in the summer. Many customers visit stores regularly, often just to take a look around, and expect to see new and exciting styles during the season. On top of that the assortments in non-food are often much larger than in the food sector, this is especially true for large department stores (as discussed in Chapter 1). A result of changing fashions is that many SKU's are only bought once. An order is placed with a supplier delivered to a store and then sold down to the last piece. In combination with the seasonal nature of fashion this means it only happens relatively infrequently that certain items are allocated a fixed amount of space. Where supermarkets use space management (planogram) software to plan and control the allocation of shelve space down to the last centimeter, fashion retailers generally don't have figures on the amount of space certain styles are getting on the floor. The amount of space an article uses is an important factor in the calculation of Direct Product Costs within the store. When information on the whereabouts of a product (in storage or on the floor) and the amount of room it is utilizing is not available, only a limited number of cost components can be allocated. The next chapters describe C&A Germany and the way a cost accounting system was set up there during the 1980's that allocated direct product costs to different classifications of product. This is followed by an explanation of the Assortment Model, which provides insight into four components which together determine a retailer‟s gross margin income. 18-6-09 34 Chapter 3 Description C&A formula in Germany This Chapter is intended to provide some background on C&A Germany to better understand the unit cost research described in Chapter 4. It starts with a short description of the company's history and expansion. A global overview is given of the organizational structure, but the focus is on the principles used to plan and control the flow of goods through the stores. 3.1 Company history C&A is a privately-owned company retailing men‟s, women‟s and children‟s clothing from department-style stores and is Europe‟s largest clothing multiple (Management Horizons, 1992). C&A was originally set up by two brothers, Clemens and August (hence C&A) Brenninkmeijer. They were in their early twenties and started on a loan of ten thousand guilders supplied by their father. At the time the north of Holland was economically more prosperous then the area where they came from, North-Westfalia in Germany, in particular the town of Mettingen. They set up their business in Sneek, Friesland. For a long time they sold goods, especially linen from their home town, door to door. The first real C&A shop which they opened in 1861 was a moderate success. Their sons took over the business and it slowly grew with stores opening in Leeuwarden and Groningen. In 1908 they started stores selling confectionery, clothing already made instead of just materials or clothing made to order. C&A was one of the first companies to make good ready-towear clothes available to ordinary people at an affordable price. By 1910 C&A had established a large chain of shops throughout Holland. Expansion abroad was the next step. Family members of the third generation founded C&A companies in Germany in 1911 and in England in 1922 (www.c-and-a.com). After the war another wave of expansion took place. In the United States stores under the name C&A existed shortly in the 1950‟s and were replaced by chains of other retail outlets which were bought as going concerns. Further European expansion started in 1963 with a chain of stores in Belgium. In 1972, C&A opened in France, in Switzerland in 1977, in Luxembourg in 1982 and in Spain in 1983. Austria followed in 1984, Portugal in 1991, Denmark in 1995 and the Czech Republic in 1999. Expansion also took place in South America in Brazil under the C&A name and 18-6-09 35 using the C&A subbrands and positioning. More recently stores have been opened in Argentina and Mexico. In the 1980‟s eleven stores were opened in Japan with a bit more of an upmarket positioning, but were closed due to lack of financial success. The assortment at first consisted solely of ladies‟ clothing, mainly coats and dresses, without undergarments. Men‟s departments were added in the same stores which was a novelty at the time. Children‟s collections rounded out the assortment making a complete offering for the whole family. Clothing lines were later extended to include undergarments and lingerie. Ladies shoes and accessories such as bags and nylons were added in the late seventies and early eighties. Men‟s shoes as well as children‟s were also added later as fur coats were dropped from the merchandise mix. The focus of this chapter is research done at C&A Germany between 1980 and 1990. During that period the German organization was allowed to operate fairly autonomously in relation to the C&A companies in other countries. A leading information source on the retail sector summed it up as follows, "A key factor in C&A's success in Europe is the fact that it adapts its market positioning to the individual markets: the company has a more upmarket image in Germany where it targets the middle market, but has a lower market positioning in the UK, for example. "(Management Horizons, 1992) 3.2 Store Positioning and Merchandising Strategy German C&A stores are quite large, some outlets have 7 floors and more than 10 000 sqm of selling area. Having stores comparable in size to department stores while focussing on apparel allowed C&A to cater to the needs of various customer segments. The mission statement emphasizes this, "A wide range of clothing is absolutely fundamental to C&A. The breadth of selection provides a competitive advantage over any other fashion retailer." In segmenting the market, C&A acknowledged differences in budget as well as in taste. Three brands were used in the past to distinguish between the price levels offered. These were „small purse‟, „happy medium‟ and „exclusive class‟. Dresses for example between 0-10 DM were „small purse‟ the price classes 10-20 and 20 and up, were respectively „happy medium‟ and „exclusive class‟. Advertisements were also set up to distinguish between these three levels of price and quality. 18-6-09 36 During the nineteen-seventies, other brands were introduced to reflect the individualizing of society. These included Westbury, Your Sixth Sense, Palomino, Canda, Angelo Litrico, Yessica, Avanti, Jingler‟s Jeans, Rodeo, and Young Collections. Besides this a number of third party brands were offered in Germany including Carl Lagerfeld jeans for women and Yves Saint Laurent for men‟s wear as well as ladies nylons. A strong brand has the ability to achieve a combination of higher selling prices and/or volume advantage over competitors selling at the same price. Albert Heijn in the Netherlands for example, is known to be more expensive than its main competitors and still maintains a dominant market share. C&A chose to focus solely on the volume premium. The C&A formula was based on a strictly adhered to strategy of 'low prices - high turnover'. This was achieved not least by being satisfied with a modest gross margin (Miellet, 1994). Prices were competition orientated, the internal guideline was that C&A's prices should be 10% under the competition. "Best value is the foundation of C&A's business. Best value built C&A and must be the area where C&A is not beaten or undersold by any competitor. Value does mean low price but is the result of the relationship between quality and price. Therefore we must ensure that at all levels of quality sold by C&A, the price is better than that being charged by any competitor for a similar quality. The realization of the underselling objective will be monitored." (C&A Mission Statement, June 1997). Everyday newspapers were scanned for advertising by competitors and employees went to other stores to compare prices and quality. If an article was found to be sold cheaper or for the same price, the comparable article at C&A was taken out of the racks and marked down against the competition on the same day. Often the very best selling articles would be marked down regardless of how profitable they were. This was seen as an investment in the brand equity of the store. No article or team was allowed to become a free-rider by having prices above the market price. Walton described the importance of consistently meeting and beating the competition on price, "Kmart really took us on in about 1977…and we fought back. We told our manager there, "No matter what, don't let them undersell you at all, on anything." I remember he called me one Saturday night and said, "You know, we have Crest toothpaste down to sic cents a tube now." And I said, "Well, just keep it there and see what they do." They didn't lower it anymore than that and we both just kept it at six cents. Finally they backed off.. I always thought they learned something about us at that store -that we don't bend easy-because they never came at us with that degree of price cutting anywhere else." (Walton, 1992) 18-6-09 37 Historically C&A was Germany‟s largest clothing retailer based on turnover. According to Eglau (1972) C&A Germany had 72 stores with a turnover of 2,7 billion DM. Among retailers including supermarkets C&A held the fifth spot in size. In clothing however it had an estimated market share of 15%, in some cities almost 35%. C&A was ten times as big as the second place specialist fashion retailer Dyckhoff. It sold more clothing than all the major department stores (Karstadt, Hertie, Kaufhof and Horten) combined. At the high point of the season a single store could sell more than 1000 ladies coats in a day and three times as many dresses. The gross margin (excluding 11% VAT) was reported to be 29% of sales and profits an astounding 10% of sales. In my opinion, C&A could be seen as a category killer in the product group clothing. Category killers, "Aim to offer the widest possible range, minimize out-of-stock, and undercut the competition by 10-25% on price. They aim for a dominant (20%+) share in the catchment area. Become the low cost operator, through a combination of low rents, low cost shop fits, tight staffing, economies of scale (labour and overhead) at store and chain level, buying muscle, and high asset productivity (space and stock turn)…Consumers are drawn to category killers on the basis of price and range authority. They know they are unlikely to find a cheaper store, that the store is likely to carry what they want, and what they want is likely to be in stock"(Wileman & Jary, 1997). This formula has worked well across a wide variety of product categories, retailers such as Toys 'R' Us and IKEA are leaders internationally as well as in their domestic markets. It can also be argued that C&A Germany's culture was a determinant for its success and innovative cost accounting systems. Almost all training was done internally, most managers were recruited out of high-school and learned on the job during a six year apprenticeship in which they lived in a different city every year. This was done to learn about the different parts of Germany and also so that the different people working in the company would know each other personally and trust on another. C&A did not feel inhibited by rules or conventions of the retail trade. During the early 1920's C&A used airplanes to counter the effects of hyperinflation. Daily cash receipts were flown to the Reichsbank in Berlin where they were exchanged for Dutch guilders. Speed was always a high priority at C&A where stock turn is known as TOS (Turn Over Speed). “At C&A England it is not unusual for a buyer to get an idea in Japan or California, then develop a new style, arrange for production and delivery and have the finished goods in the store within 3 weeks”(Eglau, 1972). People were always ready to try new things such as selling a shirt and necktie together in one 18-6-09 38 package. Shoe manufacturers were ask to deliver shoes on hangers to ease handling and transport to the stores as well as eliminate the nuisance of dealing with shoeboxes. 3.3 Implementing Merchandise Strategy 3.3.1 Organization for Merchandise Management In Chapter 1, Van der Kind's retail value chain was compared to Porter's generic value chain. One of the main differences is that in retailing procurement of goods is considered a primary activity not a support activity. Procurement in retailing is commonly referred to as buying. Generally operational activities in retailing are divided into two sectors, store operations on one hand and buying and merchandising on the other. Store operations entails responsibility for human resource management in the stores, building and fixture maintenance, relationships with the local community and other activities not directly connected to the merchandise mix. The buying and merchandising functions are responsible for the merchandise mix, that is the goods flow through the stores. Activities include demand forecasting, budgeting, setting up range, vendor selection, vendor negotiations, ordering, pricing, allocation of goods to the stores, markdowns and clearance management. The goal is to implement the firm's overall merchandise strategy in the most profitable manner. In the past retailers worked with decentralized store level management and decision-making, where store managers made the majority of the buying and merchandising decisions (Wileman & Jary, 1997). One person was in charge of store operations as well as the merchandise mix. This often led to problems, because, it is usually easier to find a good buyer capable of assuring the purchases of several different departments, than to find a buyer who can at the same time successfully direct the sales force and inspire sales promotion. ( Pasdermadjian, 1950) Several department store chains in Germany such as Karstadt still place much buying and merchandising authority and decision making in the hands of individual store managers. The majority of large retail businesses in Europe and North America however, have moved on to the centralized organizational model, with authority centered around a central buying and merchandise (B&M) function. Usually, B&M is organized around categories or product groups. Within a particular category the senior B&M manager is the buyer. Underneath the buyer are buying assistants and merchandisers (Wileman & Jary, 1997). 18-6-09 39 The organizational structure at C&A differed from the centralized organizational model described above. Where other companies were split into two main parts: centralized buying and merchandising on one hand and store operations on the other, in C&A Germany three management functions stood on an equal level in the hierarchy. At C&A a split is made in store operations, merchandise management and buying. This organization was set up at the time that buying for the whole of Germany was centralized in Dusseldorf around 1969-1970. The merchandising function was separated from the buying function, merchandisers were primarily responsible for a division in a geographic region, not for a single classification of clothing for whole country as the buying teams were. The merchandise managers worked and lived in different areas around the country, not in the head office. They were responsible for allocation of goods to the stores, presentation, space management, markdowns and clearance. The necessary analytical skills and the capacity to work with the store employees and management did not completely overlap with those of the buyers. The ability to set up a range with a strong style point of view, find suppliers (domestic and overseas) and negotiating buying terms, oversee product delivery and manage overall stock levels were prerequisites for buyers. Nevertheless many buyers spent time working as merchandise managers and vice versa which was definitely an aid to cooperation and understanding between the two functions. 3.3.2 The buying function The buying function was brought under a legally separate company named the Eteha Textielhandelsgeselschaft. Both the buying and merchandising functions were divided into 6 divisions with roughly equal revenues: Ladies Coats & Dresses, Ladies Fashion, Ladies Various, Men‟s Outerwear, Men‟s Fashion and Children. Within those divisions, buying teams were split up 18-6-09 40 along narrowly defined parts of the assortment called classifications. For the commodity dresses for example there were 5 teams; one for dresses up to an original retail price of 29,75 Marks, two for price levels between 30 and 59 Marks and two for the price levels above 60 Marks. In the medium and high price levels a distinction was made between young, fashionable styles and more conservative dresses. It is unusual for a retailer to divide buying amongst low, medium and high priced teams for a single commodity. These teams competed directly against one another for sales on the same selling space. Buying was done by each team individually, there was little co-ordination between teams. The collections were put together in secrecy, new collections would be hidden behind curtains to prevent others from seeing what was planned for the coming season. Teams responsible for the same commodity did not show other teams what they were buying or at what cost. Some buyers even threatened their own suppliers that they would stop working with them if they sold to other C&A buying teams. It was not unusual for a buyer to visit all the links of the retail value chain in a week; a supplier's factory, the head office, a distribution center and a store. Every Friday buyers would return from factory visits or work in the central buying offices in Dusseldorf to their hometowns around Germany to meet merchandise managers in the stores and discuss the collection, trends and the competition. This ensured a steady communication flow between the different regions of the country and buying organization. David Glass, former Wal-Mart CEO described the importance of similar meetings for his company: "Our Friday merchandise meeting is unique to retailing as far as I can tell. Here we have all these regional managers who have been out in the field all week long they are the operations guys who run the run stores. Then you have all your buyers back at headquarters. In retailing there has always been a traditional, head-to-head confrontation between operations and merchandising. You know the operations guys say, "Why in the world would anyone buy this? It's a dog, and we'll never sell it." Then the buying folks say, "There's nothing wrong with that item. If you guys were smart enough to display it well and promote it properly, it would blow out the doors." That's the way it is everywhere, including Wal-Mart. So we sit these folks down at the same table and just have at it" (Walton, 1992). 18-6-09 41 3.3.3 Merchandise Managers and the flow of goods The merchandise managers for there part, as mentioned above, were responsible for a division of goods (such as children's clothing) for stores in their region. They had to make sure the allocation was running smoothly and stocks per store were at the correct levels. They were responsible for the presentation and keeping an eye on the competitor's offer in their part of the assortment. On top of that, each merchandise manager had an extra responsibility for one or two commodities for which he was a contact or liaison between merchandise management and the buying teams concerned. Merchandise managers were responsible for the flow of goods through the stores. C&A has always defined retailing more in terms of moving goods than in presenting and then selling them. Where normally in sales analysis, categories are called fast sellers or slow sellers (Ghosh, 1994), at C&A they are referred to as fast and slow movers. Stock turn is referred to as turn over speed. Stock turn has always been seen as the most important factor for controlling merchandise at C&A Germany. Every store had approximately the same turnover speed because goods were allocated according to the speed at which the goods were sold in relation to the amount of goods in the stores and reserves. This was done on classification and on code level. This meant that there was no direct relation to store selling area and the amount of clothing allocated to a store. So one store might have three times more clothing allocated to it that it had floor space while another store might just be able to fill it‟s selling area. This was accommodated for by the use of reserves in the distribution centers, where clothing was stored for individual stores. 18-6-09 42 Figure 5 Stock turn based inventory allocation within a classification New goods Distribution Center Total stock for store determined by sales, not by selling area DC Store A Store B Equal store size: same stock turn, lower sales per sqm Sales The figure above represent the flow of goods through two different stores with similar amounts of selling area. The blocks each represent an amount of stock, the total stock allocated to each store is related to the amount sold each week. In general new clothing arrived daily from the distribution centers. Each center could function as a receiving point for supplier deliveries, a distribution point to the other distribution centers and a supply point for stores in the area as well as a short term storage area for goods not needed immediately in the regional stores. The usual practice for a new style of clothing was that a single standard unit (pre)pack with a range of sizes would be sent to the store and the other packs allocated to that store would be stored at the distribution center. The merchandise manager could then decide depending on the situation in the store when to call off the rest of the packs. This happened in one or more allotments of standard packs, not by individual piece. The new styles were presented along the walking paths. The fixtures used for this were made up of a number of arms or shelves set up to show different style or colors together. For example a fixture could have four arms in a cross form displaying two styles of blouses in two different colors. These styles were at first available in all sizes. As they sold out and new merchandise arrived, older styles would be put into the racks which were arranged to size. New styles came to replace the styles which were partly sold. In the size rack the number of pieces presented per size was in proportion with the sales in that particular size. 18-6-09 43 Figure 6 Trouser fixture seen from above Figure 6 illustrates some Size 50 48 46 44 of the reasons why it is difficult to determine the amount of space used by a fashion article, which is one of the fundaments of DPP. The first is the Size 52 New goods 54 56 58 constantly new style changing will be styles. Where and how a presented depends largely on the situation in an individual store at the moment it arrives. Another reason it is difficult to allocate space to a certain style is that the number of articles decreases as it sells. Finally a single style can be presented to size and thus spread out over a fixture amongst other styles. Figure 7 C&A stock turn system At C&A Germany, sales area was allocated to different commodities such as trousers. Within a commodity low priced and high priced articles were presented next to one another on the same racks. Management realized that GMROI did Stock turn Stock turn in relation to price (within a commodity) not have to be equal across price points, because a higher price represents a higher inventory investment (as stated in section 2,2). Articles had to achieve similar gross Low Medium Average price High margins percentages, but a difference was made in stock turn targets depending on the average selling price of a classification. Within a commodity the lower priced goods were given a higher stock turn target than the high 18-6-09 44 priced goods. This ensured that each classification within a commodity had the same gross margin per square meter (GMROS). A pair of pants with a gross margin of 10 DM has to be sold twice as often has one with a 20 DM gross margin to achieve the same return in the same area Markdowns were done for two main reasons, firstly and most importantly if goods were not selling fast enough. Markdowns were taken very quickly, based on sales figures for styles at the end of the week. A slow-mover doesn‟t usually start selling any faster by itself. According to German law it used to be forbidden to offer customers a lot of marked down prices during the season. The socalled “Karenzseit”, therefore markdowns were done invisibly. Tickets were taken off and new tickets were attached with the new marked down price without any mention of the old price. The customer couldn‟t see whether a particular product had been marked down. This also had the added advantage that the customers didn‟t have the feeling that they should wait for markdowns or that the normal selling price was to high. Too many visible markdowns will have a negative influence on a retailers image (Buuron, 1995). At any one time there were very few goods in the store which customers considered too expensive in relation to the fashionability and quality of the article because slowmovers were constantly weeded out and lowered in price. The end of the season clearance sales were used to clear out all remaining goods brought in during the past season with the exception of basic or staple items which were replenished throughout the year. None of the articles in the stores were returned to central storage to be brought back to the sales floor a year later. Goods were marked down progressively until everything was sold out. The very last goods deemed unsellable were written off. The important thing to consider was that the focus was on income per square meter of selling space and not on the return on an individual piece of clothing (as was discussed in the paragraph above). Goods which are not sold, block the selling space so there is not only the depreciation of the goods to consider and capital tied up in merchandise. Slow movers on the floor also represent an opportunity cost, that is the opportunity to sell other goods in the same place. 3.3.4 Store operations In the store the general practice was to display goods in such a fashion that handling costs were minimized and selling maximized. Different fixtures were used to present the clothing including 18-6-09 45 tables on which goods were dumped unceremoniously. The store was set up to enable self-serve shopping, none of the sales staff received commission and a split was made between sales and cash desk staff which will be discussed in Chapter 4. To achieve financial goals each division and each classification of goods within a division had to achieve a certain level of sales per square meter. Floor area was divided up amongst the different divisions based on their productivity in the preceding months. If the children's division had the highest turnover per square meter in a certain store, it would receive an increased proportion of the sales area during the following period. The carpets in the selling area had dots every 10 cm's to help measure which racks could be placed where. This meant that the space allocated was an indication of the sales achieved. Exceptions were made, for example when new articles were introduced, they usually got more space then turnover justified for a number of years to increase market share. Sales per square meter is an estimate of return on investment only as long as two criteria are met. The first condition is that the margins used do not vary considerably. This was generally the case at C&A. The second is that the costs associated with handling the goods within a certain area are not very different from the costs of handling per square meter across the whole store. These selling, general and administrative costs were the focus of an extensive research program carried out at C&A Germany during the 1980's which will be discussed in the following chapter. 18-6-09 46 Chapter 4 Unit Cost Research and the Assortment Model In this chapter the selling process and articles sold are approached from two different perspectives. The unit cost research carried out at C&A focussed on the costs involved in selling clothing especially costs incurred in the actual stores. The Assortment Model, which is discussed in the last section, is not concerned with costs but with the components of gross margin. 4.1 Cost Accounting Research Method C&A Germany This section looks at the links between C&A and some of the factories that supplied it with merchandise. The differences between manufacturing and retailing are discussed from a cost accounting point of view. The research scope and objectives are presented followed by an illustration of the method using examples. Warehouse and transport costs are covered, but the most attention goes to store costs, which in practice make up a large percentage of costs but are seldom analyzed in detail in fashion retailing. Finally a comparison is made between the unit cost research and conventional measures of retail success as well as Direct Product Profitability. 4.1.1 Factories as an image of store organization In the 70's approximately 7 500 people worked in 23 different European clothing factories owned by the Canda (C and A) company (Miellet, 1994). In Germany the factories produced: coats and suits in Ludwigshafen, dresses in Essen, children‟s coats in Wattenscheid, knitwear in Neu-Ulm and men‟s wear in Mettingen, Esterwegen, Freeren and Haselünne (Eglau, 1972). Most C&A executives were expected to spend at least 6 months working in a factory as part of their training. Around 10 and 20% of the clothing sold in the stores were produced in these factories. Knowing the ins and outs of production and its costs gave C&A buyers an extra advantage when dealing with other suppliers (Eglau, 1972). It can also be argued that in store operations C&A management paradigms concerning revenues and costs were more closely related to manufacturing than conventional retailing. 18-6-09 47 Companies that supply goods to retailers (often manufacturers, who are limited by technical restraints of their production facilities to a limited number of similar articles) will tend to express revenue (sales) as the product of price multiplied by the number of units sold. Sales= Price x Units sold According to Van der Kind however, the approach is less meaningful for retail, "because one of the most important facets of retail is that companies operate in many different product categories simultaneously. It is therefore impractical to calculate sales for each separate product. That is why assortiments used in retail consist of a certain number of SKU‟s together. Often such assortments are bought by one buyer and for him his turnover is expressed as: Sales = marketshare x marketsize " Figure 8 The Revenue Triangle (adapted from Van Kind, 1996) Manufacturer Sales = Price x Units sold Store operations Sales=Customers x Amount Spent (C&A: Sales/square meter ) Buying & Merchandising Sales = Marketshare x marketsize (C&A: Sales = Price x Units sold) For buying and merchandising the difference in viewpoints concerning revenue is reflected in the methods used to divide the assortment into parts. At C&A the commodity split into low, medium and high prices and the different stock turn targets for each price range to reach comparative revenues per square meter reflected a manufacturer's viewpoint. Normally in retail, a classification is an assortment of items that are regarded by the customer as being reasonably substitutable for one another when she is buying for an end use (Buuron, 1995). In this case, the assortment is approached from a market viewpoint. 18-6-09 48 Table 7 Examples of assortment breakdown (National Retail Federation taken from Buuron, 1995) Division NRF 300 Adult male apparel and accessories C&A M999 Men's Division Department 330 Men's Casual Wear M63 Men's Trousers Classification 334 Casual and washable slacks 371 Trousers Low 372 Trousers Medium 373 Trousers High Code Cotton Knit Cotton Wool For store operations it must be said that, purchase per customer has long been known as a good indicator of cost structure. The larger the purchase per customer, the greater the effectiveness of the retail operation. The number of customers that have to come into the store to move a certain quantity of a certain item of merchandise may well be a more reliable unit of transactions (to calculate costs) for stores than any other (Drucker, 1963). The success of C&A was based largely on a consistent cost consciousness (Eglau, 1972). To measure costs and streamline processes in store operations scientific management techniques were used in special test areas in Dusseldorf. Close links with factories meant that they would be more inclined to use things like time and motion studies then people who have worked exclusively in retailing. Cost studies have lead to many innovative ways of doing things such as reusing hangers and having price tickets attached to clothing at the supplier where costs of labor are often lower. Time and motion studies have determined the layout and procedures for such essential processes as the checkout. This industrial way of thinking was probably also instrumental in the decision to consider individual articles of clothing as cost objects for which realistic costs should be calculated. A parallel can be drawn between consumer products producers such as General Foods and P&G introducing Direct Product Profitability to supermarkets and the transfer of cost accounting methods from the Canda factories to the C&A stores. Historically, accounting techniques for planning and control arose in conjunction with manufacturing rather than nonmanufacturing, because the measurement problems were less imposing and environmental factors such as economic conditions, customer reactions, and 18-6-09 49 competitor activity were generally less influential in manufacturing (Horngren & Foster, 1991) . In a typical department store in the Netherlands for example there are moments that stores can be up to a thousand times busier than others, say the Saturday before Christmas versus a Tuesday morning in February (Van der Kind, 1996). The basic concepts of planning and control however, apply equally well to both manufacturing and nonmanufacturing activities (Horngren & Foster, 1991). A more fundamental reason for the historic lack of cost accounting is that retailers are not involved in providing “form” utility as opposed to manufacturers (as discussed in Chapter 1). Merchandising is the marketing of goods without changing their basic form. A manufacturer on the other hand, transforms direct materials into salable form with the help of direct labor and factory overhead. The rules of financial accounting have a major influence on accounting for manufacturing costs. Under generally accepted accounting principles, the manufacturing costs of a product are inventoriable costs. A manufacturer traces costs to make them a part of inventory. Such costs become expenses (in the form of cost of goods sold) only when the units in inventory are sold. Such sales may occur in the same accounting period as manufacture or in a subsequent period. In retailing all sales, general and administrative (SGA) costs are considered period costs. They are regarded as immediate expenses, they are not assigned to inventory. The only inventoriable cost is the cost of merchandise, i.e. the amount paid to suppliers. Unsold goods are held as merchandise inventory whose cost is shown as an asset on the balance sheet. As the goods are sold, their costs become expenses in the form of COGS (Horngren & Foster, 1991). 4.1.2 Research Scope and Objectives The research into the costs of selling different types of articles at C&A Germany was known as the unit cost research (Stückkostenuntersuchung ) The research was developed and carried out during the years 1980 until 1991. The market research department was responsible for the methodology, data collection and processing. The results were tabulated and presented once a year. The main goal was to get accurate results on the cost of selling articles for the smallest possible cost object. A cost object is any activity or item for which a separate measure of costs is desired. Normal cost objects for retailers are organizational departments such individual stores, the head office, buying and distribution facilities. The cost object in Direct Product Profitability is the stock keeping 18-6-09 50 unit (SKU), for example Calve peanut butter in a 300 gram jar. The cost objects in the C&A research were individual classifications, such as all umbrellas or low priced men's shirts. The entire assortment consisted of around 225 classifications. The costs per unit were calculated by dividing the costs allocated to a classification by the number of units it sold. Five stores of differing sizes, that were meant to offer a cross section of all the stores, were used to collect information on the costs of selling staff. Only about 40% of the total selling, general and administrative (SGA) expenses were included in the research. Expenses that were not included are summed up in the following table. In general, they are the expenses not included in the store management or physical distribution budgets. Table 8: Scope of the unit cost research (1983 Numbers in millions of DM) Gross margin Allocated SGA expenses 660 Not allocated Store Rents 341 Taxes 132 Advertising 147 Interest 40 (Head office) overhead 244 Buying organization 126 Total SGA expenses Profit 2137 1690 447 In the above table, it is especially interesting to note that the rents were not included in the calculations. The amount of space per classification was measured, but rent was kept out of the calculation while occupancy costs such as heating, electricity and cleaning were all allocated based on space used. The unit cost research was carried out for a number of different reasons. The purpose of the research was not external or fiscal reporting. It was believed that a focus on classifications of articles as cost objects would be an aid in decision-making. Cost information can be relevant for questions such as (Horngren &Foster, 1991) Which products should we sell? Discontinue? What prices should we charge? Should we change our selling methods? 18-6-09 51 Which products should we sell? Discontinue? The addition of new articles was determined to a large extent on plans based on different aspects of the unit cost research. The separation of costs into parts facilitated the forecasting of the unit costs of possible new articles. For the addition of shoes, lingerie and nylons to the collection, costs of handling in the store and distribution centers were estimated before their introduction into the assortment. Articles with very low prices were either dropped or sold in units made up of more than 1 piece or pair. Markdowns were also done to a minimum level of 3,- DM, it was not considered worthwhile to have prices any lower than that. What prices should we charge? C&A does it planning in the conventional manner, using a single company wide gross margin percentage target, which is to be achieved to cover SGA expenses. In 1983 the market research department wrote: "(The unit cost research) is a prerequisite for a differentiated approach to the market. This is particularly relevant in today's market where increasingly new competitors are appearing, that are characterized by price aggressiveness." It is not clear, whether this meant using different margins depending on the competition for certain parts of the assortment or offering extra service for some product segments to legitimize higher margins than competitors. Should we change our selling methods? Cost information was the basis for internal efficiency improvements in store operations. Decisions were made not to sell sweaters on tables but on hangers to cut down on the labor involved in keeping them presentable. The amount of time between orders for nylons was doubled and the stocks increased (lowering stock turn) to make handling more efficient. The calculation of unit costs on classification level had a central to a long-term character due to its methodology. The calculation was only carried out once a year and reflected historical figures. It could be used as a guide for strategic planning. For short term, daily decisions within the store, applicability was limited because the results were averages over different stores and different periods of time. 4.1.3 Broad description of the research In cost accounting, costs are considered to be either direct or indirect. Direct costs are those costs that can be identified specifically with or traced to a given cost object in an economically feasible 18-6-09 52 way. Indirect costs cannot be identified specifically with or traced to a given cost object in an economically feasible way. In general many more costs are direct regarding a department as a cost object than a physical product (Horngren & Foster, 1991). Costs of materials as well as direct labor have traditionally been the direct costs associated with manufacturing. In most retailing operations only the amount paid for merchandise (Cost of goods sold) is a direct cost. All SGA expenses are considered to be indirect costs. The essence of DPP and the unit cost research is to make SGA expenses direct costs, which was (economically) unfeasible in the past before the advent of inexpensive computing. Both direct and indirect costs can also be categorized as either fixed or variable depending on the way they react to changes in the cost driver. A cost driver is any factor whose change causes a change in the total cost of a related cost object. Variable costs are dependant on changes in the cost driver, for example 1 paper price tag is needed per article sold. Fixed cost are unchanged over a wide range of the cost driver during a given time span. They become progressively smaller on a per unit basis as the cost driver increases in the relevant range as the fixed costs remain constant. The total labor costs in a store for example, will not decrease if the employees reattach 800 price tags instead of 1000, but the cost per tag would increase. While some authors such as Coopers & Kaplan (1988) encourage companies to regard all their costs as variable, it is important to realize that the great majority of costs in retailing are fixed in the short term. Changes in volume will affect variable costs but not total fixed costs. Unit costs are useful, but must be interpreted with extreme caution, especially if they are in the form of fixed costs per unit (Horngren & Foster, 1991). When nylons were introduced at C&A Germany for example, some people argued that they were an inherently unprofitable article due to the required handling and relatively low selling price. At first the unit cost research seemed to support their argument, but when the sales of nylons increased, the unit costs for the identical article in the same stores with the same employees, etc. told a completely different story. The minutes of time spent tending to the nylons and the amount of space used per unit dropped as more units were sold. The research covers the entire goods flow, from the receiving point at the distribution center to the checkout counter in the stores. The total costs of the individual departments, taken from the store 18-6-09 53 operations accounting system, divided by the total number of units sold results in the average cost price per unit for the whole company. Table 9 Department costs as the basis for calculating unit costs (C&A Germany, 1983) Cost pools Warehouse costs - Labor - Overheads Transport costs Store costs -Labor Sales staff Cashdesk/Checkout Other store staff -Overheads Mil. DM 50,2 10,7 38,1 247,2 86,3 94,8 109,6 Cost application bases to products sold Time Time, units, delivery condition, volume, order size Volume in cubic centimeters Time (Multi Moment Recording) Time (Time and motion studies) Space, value, units, sales staff time Space, value, units, sales staff time Total (in Mil. DM) 660,3 Divided by total number sold 1983: 203 892 250 units sold = average unit cost 1983 = 3,24 DM/unit In order to find the required unit costs per individual classification, for each activity (transport, ticketing, selling , checkout, etc..) methods to be developed, in order to determine the relevant labor and material costs. The results of measurement, for example man minutes per unit for a certain classification, are weighted with the annual number of units sold, in order to show the actual results for a total year. The total result was the sum of the results of all individual parts of the research, which were executed in different ways and independently from one another. The chance of making major mistakes was thus reduced. The individual parts of the research are each based on the costs of the departments shown above taken from the financial accounting system. Thus, the store measurements could be done at different times and independent of the distribution center. This allocation of the costs to the classification was executed independently for each part of the enterprise, for which a separate calculation is meaningful. This simplified subsequent profitability analyses carried out over a number of years. 18-6-09 54 4.2 Illustration of the method using examples To give the reader an idea of what the unit cost research entailed especially in the store environment the buildup of the research will be illustrated below using 5 classifications of goods as examples. These are high priced ladies coats, men's ties and men's trousers in the low, medium and high price ranges. The first, high priced ladies coats, is a big-ticket item traditionally seen as one of the mainstays of the C&A collection. Men's ties had the reputation within C&A of being the single most profitable article based on the results of the unit cost research. Men's trousers were also used as an example in Chapter 2. The differences between the three classifications of trousers are also indicative for other articles sold at a wide range of price points. Of crucial importance for the success of the research is the careful analysis of the single activity areas, where research methods have to be developed, tested and finally correctly applied. The individual parts can be quite complex, the following overview just gives an impression of the methods used for the individual departments. The absolute calculated costs per unit in 1990 are given. It is assumed that each classification has to carry the share of the costs which is allocated to it, there were no loss leaders at C&A.. The cost areas that were part of the research are listed below with the relevant cost drivers and the results for each of the 5 examples. The list is roughly arranged to follow the flow of goods from the distribution center to the cashdesk in the store. The type of the cost driver used is either direct through an objective measurement or an indirect cost application of overheads based on the departments with the closest connection to it. 4.2.1 Warehouse costs The calculation of the costs in the distribution centers is not explained in this paper. There is already a lot of detailed literature on costing for warehousing and distribution centers. The measurement problems involved are different and less daunting than in store operations and are more comparable to manufacturing environments where factors such as customer reactions and competitor activity are less influential. Finally, the costs allocated amount to less than 10% of the total involved in the unit cost research. 18-6-09 55 Twelve different inputs per classification were used to calculate the cost drivers involved in the distribution centers. These included amongst others the number of orders/deliveries, delivery condition (hanging or in boxes), the volume per article and quality inspection requirements. Detailed time and motion studies were carried out for different possibilities for all operational sequences. Distribution Center Costs per Unit Warehouse Offices Cost per unit 4.2.2 Transportation costs Transport costs in this case are the costs involved in transporting articles from the distribution centers to the stores. All goods were delivered through the distribution centers, there was no direct store delivery. C&A Germany outsourced transportation to a third party distribution company that was paid based on the number of truckloads transported during a year. An increase in volume thus led directly to an increase in transportation costs. The cost driver was therefore the average volume in cubic centimeters of the articles sold within a classification. The volume was determined by a special investigation of the size per piece for each classification. Here it concerns average values, the volume of a single article can be different. It is well known for example that the volume of a winter coat is larger than a coat sold in the summer. The extensive study to estimate the average volume per unit for each and every classification was not carried out annually. The results of this type of measurement were used over a number of years to allocate (transport) costs. Per classification a total volume was calculated by multiplying the quantity sold by the average volume per unit. By summing the values of all classifications a total company volume that was transported resulted. This total calculated volume was related to the total transport costs in the year concerned. In 1990 the total company volume resulting from the sums of the calculation was 1 371 634 156 226 cubic centimeters. The sales quantity was 262 099 089 units. Total transport costs amounted to 65 476 785 DM. 18-6-09 56 Average 0,26 DM Average 0,12 DM Lad. Coats H Men's Ties 0,82 DM 0,07 DM Lad. Coats H Men's Ties 0,65 DM 0,08 DM Trousers L 0,24 DM Trousers L 0,13 DM Trousers M Trousers H 0,34 DM 0,89 DM Trousers M Trousers H 0,20 DM 0,36 DM Transport Average Volume per unit 5 233 ccm Cost per Unit 0,25 DM 4.2.3 Store costs Lad. Coats H Men's Ties 26 358 ccm 749 ccm 1,26 DM 0,04 DM Trousers L 10 859 ccm 0,52 DM Trousers M 13 171 ccm 0,63 DM Trousers H 13 464 ccm 0,64 DM The store costs included in the research and discussed below are labor costs for sales and cashdesk staff, shrinkage and store overheads. Sales staff labor was responsible for about 40% of the allocated costs. The sales staff are the people who work on the sales floor helping customers and taking care of the merchandise. Compared with all other areas covered by the research the largest difficulties in methodology had to be overcome here. It is not enough to just measure the costs of moving stock to the shelf or rack as is customary in DPP calculations for the food sector. Much of the time is spent helping and advising customers as well as tidying the merchandise, changing the presentation and doing markdowns. A procedure was worked out through trial and error that was considered informative, reliable and not too obtrusive. It is known as Multi Moment Recording. In a number of stores for limited periods of time, every member of the sales staff was asked to fill in a standard form recording their activities during the day. The procedure worked as follows. Every five minutes a bell sounded in the store, employees would then note the classification of the article they were occupied with at that moment as well as the activity concerned. Separating the staff labor costs into the different activities enabled further analysis of the cause of unit cost differences. A large number of activities were identified, defined and then grouped into zeven main categories: 1.Customer contact 2. Stock movement 3. Tidying merchandise 4. Carrying out markdowns 5. (re)ticketing 6. Miscellaneous 7. Breaks The number of the activity and the classification were written down on the form in the appropriate time slot (see Appendix). The test stores where Multi Moment Recording was first carried out were Karlsruhe, Mulheim, Nurnberg, Bochum, Koblenz and Wandsbek. They represented a cross section of different stores sizes and traffic. Trials were held in the months March, May, September and November, each time covering a period of four weeks. 18-6-09 57 The number of times a classification was recorded compared to the total number of recordings indicated the amount of time spent on that classification. The number of units sold in the test stores during the relevant time periods, were related to the required time and the minutes spent per unit were calculated. As was the case with the volume per article measurement for transport costs, the time spent per unit was not calculated annually. In fact, in 1990 data from the 1981 store measurements was still being used. While the number of minutes per article was reckoned to be unchanged, rising labor costs and inflation meant that the average cost per unit had risen from 1,16 DM in 1981 to 1,55 DM in 1990. Sales Staff Minutes per unit Cost per Unit Average 2,56 1,55 DM Lad. Coats H Men's Ties 25,62 0,85 15,47 DM 0,51 DM Trousers L 3,95 2,38 DM Trousers M 5,18 3,13 DM Trousers H 6,89 4,16 DM There is a large difference in the time spent per coat and per tie. The reasons for this are evident in the time spent for each activity for the two classifications. Customers who are spending over 300, DM for a coat will tend to ask for advice. Coats also often have to be altered for length and this has to be measured and written down. Of course for stock movement, ties are easier to handle and less time is needed for tidying because ties do not have to be sorted to size. The results can be seen in the table shown below. Table 10: An example of Multi Moment Recording results Minutes spent per unit sold Lad. Coats H Men's Ties 1.Customer contact 17,1 0,32 2. Stock movement 3,0 0,29 3. Tidying merchandise 2,6 0,06 4. Carrying out markdowns 0,3 0,04 5. (re)ticketing 0,1 0,01 6. Miscellaneous 3,6 0,12 Total number of minutes 25,6 0,85 Cashdesk and checkout: When a customer had found something she wanted to buy, she would proceed to the cashdesk. A separation was made between selling staff and cashdesks. Cashdesks were meant to be fully staffed and as productive as possible. Lines were considered a nuisance which the customer should accept in exchange for the low prices offered. C&A's philosophy here was the same as Aldi's, 'low prices are 18-6-09 58 only possible through low costs.' (Brandes, 1998). In fact, the head office, which was responsible for rules and procedures, defined a line as a group of more than 7 people waiting at a cashdesk. The result was that the people working at the cashdesk were constantly occupied. This led to high productivities but also meant that the Multi Moment Recording system was impractical for this area. Each cash desk usually had three people staffing it. They worked side by side and the customer was expected to walk past them from left to right. The first person was responsible for accepting the goods, taking off any security tags and positioning the price tickets so that they could be scanned. The second person was the cashier who scanned the price tickets and received payment from the customer, The third person at the cashdesk folded the goods and placed them in the plastic C&A bags. The cost driver was time expended per unit in seconds. To measure the time needed per classification, techniques similar to those used in factories were used. Extensive time and motion studies were carried out with consideration of multiple purchases (articles such as nylons are seldom bought separately). The results are shown below, the time used is expressed in seconds not minutes. The extra time needed for expensive trousers may be due to the fact that expensive goods were more likely to have security tags attached to them. Cash + Checkout Average Seconds per unit 29,6 Cost per Unit 0,57 DM Lad. Coats H Men's Ties 32,2 26,6 0,62 DM 0,52 DM Trousers L 28,9 0,56 DM Trousers M 36,3 0,70 DM Trousers H 39,3 0,76 DM Space used was the last aspect of the research to be measured accurately. During the first years of the unit cost research only the amount of space used by a total article group was measured, for example men's trousers and no distinction was made between the individual classifications. The reason for this was that many of the classifications were presented together with others. Low, medium and high trousers for example could all be found on a single rack (as illustrated in Chapter 3). The first measurement of actual space used per classification was carried out in 1984. It was done five times a year in a number of stores. The method of allocating space to products that were mixed together in the same selling area took place in three steps. First the amount of space for an article group (for example all ladies coats) was measured. This could be done based on the number and positioning of the fixtures concerned. These were drawn on blueprints of the store layout and the amount of space used could be noted. Secondly a count was done of the number of articles 18-6-09 59 belonging to each classification within the area allocated to the total article group. The number of units for a single classification in relation to the total number of units counted multiplied by the total area used resulted in the amount of space needed for that classification. Finally the that total amount of space was divided by the number of units sold in that classification during a year. The resulting amount of space is therefore not an indication of the amount of space a unit takes up on the floor i.e. it's density. It is a mix of the amount of room an article needs and the amount of time it is in stock. In the results below it is not surprising to see that a ladies coat needs more space than a tie. The difference in trousers however is caused by the decrease in stock turn targets as the selling price increases. The high priced trousers are in the store for a longer period of time and thus use more space. Space per unit Average Lad. Coats H Men's Ties Trousers L Trousers M Square cm per 24,2 214,1 04,76 25,49 42,17 unit per year (Total sales area at C&A Germany 1991 was 640 000 sq. m = 6 400 000 000 sq. cm 6 400 000 000 sq. cm/262 000 000 units sold per year= 24 sq. cm/unit per year) Trousers H 68,04 The figures on the amount of space used were one of the cost application bases for other store staff and overhead costs. A cost application base is the common denominator for systematically linking a cost or group of costs, such as factory overhead, to products (Horngren & Foster,1991). The subaccounts listed below were examined in detail and 4 application bases were determined for cost allocation to merchandise. For example, janitorial costs were allocated by the amount of space used by a classification because a janitor is paid to take care of the building. 18-6-09 60 Table 11 Allocation of store overheads Cost application bases 1 Sales staff; time spent per unit in minutes 2 Space; amount of sales floor area used 3 Sales value; in relation to total company sales 4 Sales in units; number of pieces sold 090 100 Office fulltime, part-time 110 120 Merchandise ticketing and administration 130 Social worker 140 150 160 Window-dressers full- part-time, training 170 180 Janitors full- part time 190 200 Cleaner full- part time 210 220 Kitchen staff full- part time 240 Training and social expenses 250 Company doctor 260 Heating fuel and water 270 Electricity 280 Cleaning outsourced 290 Cleaning materials 300 Repairs 310 Maintenance 320 ?? N. zu Inv. Geg 330 Decoration materials 340 Packaging and shopping bags 350 Telephone 370 Office supplies 380 Recruiting 390 Food and travel costs 400 Miscellaneous 4. Sales in units 3. Sales value time spent on the classification 1. Sales staff X 2. Space Here again it is interesting to note that while things like electricity are allocated according to space used, rent was not included in the unit cost research. The results of the allocation for the five examples are shown below. Average Other store staff 0,41 DM Overheads 0,40 DM Shrinkage: "As you may know, shrinkage, or unaccounted for inventory loss –theft, in other words- is one of the biggest enemies of profitability in the retail business" (Walton, 1992). On the other hand Lad. Coats H Men's Ties 3,17 DM 0,15 DM 3,43 DM 0,18 DM Trousers L 0,45 DM 0,53 DM Trousers M 0,69 DM 0,79 DM Trousers H 1,06 DM 1,14 DM Other staff Overheads X X X X X X X X X X X X X X X X X X X X X X 18-6-09 61 shrinkage is one of the easiest costs to allocate to products if detailed point of sale (POS) information is collected. C&A Germany had detailed information on exactly what was sold before many other retailers because it used computer punch cards with article information as price tickets before barcode technology existed. Skrinkage was measured directly per classification from the physical inventory/shrinkage list. The costs concerned here are mostly those of theft or loss. Goods that are damaged or unsellable are booked out per unit and not allocated to shrinkage. Physical counts of inventory were done in the stores twice a year. In 1990 total shrinkage at C&A Germany was 52 087 590 DM at cost. It is interesting to note the different shrinkage amounts in trousers. In general, the percentage of units stolen within a commodity increases in proportion to the selling price. Apparently, most thieves decide that if they are going to steal something it might as well be something expensive. It should also be noted that the stock turn for more expensive goods was relatively slower at C&A. If an article is in the store for a longer period, the opportunity to steal it is greater. Shrinkage Average Lad. Coats H Men's Ties % units 0,9 % 1,2 % 3,2 % Value per unit 0,20 DM 0,60 DM 0,07 DM 4.2.4 Editing and presentation of the data Trousers L 0,05% 0,03 DM Trousers M 0,53 % 0,23 DM Trousers H 1,81 % 1,18 DM The unit cost research was done as stated earlier for more than 200 different classifications at C&A. The data involved was quite extensive. The results were tabulated per department such as the distribution center. Within a department results were noted for every classification as well as averages for the different divisions (childrens, men's fashion, etc.). For the total company results, rankings were made according to the absolute unit costs involved in selling articles as well as the unit costs expressed as a percentage of the average selling price of a classification. 18-6-09 62 The results together with sales figures were presented as follows: Unit cost results for the 5 example classifications in 1990 Total Lad.Coats H Men's Ties Trousers L Trousers M Trousers H Sales value DM 8 040 018 683 56 614 002 66 290 388 74 026 029 131 893 905 115 674 825 Sales quantity 262 099 089 216 449 3 861 422 1 716 988 1 789 484 1 046 919 Average price DM 30,68 306,26 14,66 43,11 73,70 110,49 Total uUnit costs DM Unit costs as % of selling price 3,76 12,3% 26,02 8,5% 1,62 11,1% 4,84 11,2% 6,71 9,1% 10,19 9,2% The red numbers represent the worst result and the green numbers would be the highest ranked. In this case men's ties have the lowest absolute selling costs per unit and all the 5 articles have lower than average selling costs when expressed as a percentage of the selling price. Ladies coats get the highest ranking when costs are related to selling price 4.3 Comparison research method C&A and conventional measures The tables below contain traditional measures of retail performance for the 5 example classifications. They can be compared to the tables above. Total Lad.Coats H Men's Ties Trousers L Trousers M Trousers H Gross margin 37,2% 36,9% 44,2% 37,7% 37,8% 38,3% Stock turn 7,7 4,3 8,4 13,1 7,9 4,9 GMROI 5,3 2,9 7,9 9,1 5,7 3,6 Ladies coats have the worst result in the two most commonly used parameters of retailing, gross margin % and stock turn, which result in the lowest GMROI. The high stock turn for the low priced trousers results in its high rating. Sales per sq meter Gross margin per sq. meter GMROS 12 696 4 723 14 304 5 278 30 801 13 614 16 912 6 376 17 479 6 607 16 239 6 219 Ties are obviously the best articles when productivity per square meter is measured. The rankings for sales and contribution per square meter are the same because the gross margin percentages do not vary considerably. The next factor that can be considered is labor productivity. Lusch expresses gross margin on labor as the gross margin amount per FTE (full time equivalent employee). The unit cost research 18-6-09 63 figures do not enable a direct link to the number of FTE's involved in selling. The index of GMROL used below is based on the gross margin amount per article divided by the total labor costs in the selling area (sales staff plus cashdesk and checkout staff) allocated to that article. GMROL = Gross margin/Total store labor costs per unit Total Lad.Coats H Men's Ties Trousers L Trousers M Trousers H 11,45 DM 112,80 DM 6,48 DM 16,22 DM 27,84 DM 42,31 DM 2,12 DM 16,09 DM 1,03 DM 2,94 DM 3,83 DM 4,92 DM 5,4 7,01 6,29 5,52 7,27 8,60 Gross margin Total store labor costs per unit GMROL It is interesting to see here that the highest and lowest ranking classification are both are part of the same commodity ie. trousers. The difference in selling price leads to greater differences in labor productivity than the difference between selling a tie or a ladies coat. The table below gives an impression of the importance of unit costs as part of sales and the differences in selling prices. 300 DM 250 DM 200 DM 150 DM 100 DM 50 DM - DM Price Profit and other costs Gross margin Unit costs VAT 14% Cost of goods Average Ladies Coats High Men's Ties Torusers L Trousers M Trousers High 18-6-09 64 4.4 Comparison research method C&A and DPP There are some important differences between the unit cost research and Direct Product Profitability. These include the cost object, the methodology used as well as the activities which are included. The aim of DPP is to track the profitability of articles on stock keeping unit level. The cost object at C&A was the classification, which meant that the results were not relevant for a buying team evaluating the performance of two styles of trouser within a single classification. At C&A the starting point for the unit cost research was actual financial accounts which were allocated to cost objects. In Direct Product Profitabilty standard costs are calculated and used for different activities, so the total amount of store staff costs allocated for example does not have to be equal the actual staff costs incurred during a period. Direct Product Profitability is basically a tool for supermarkets. The assumption is made that articles only have to be put on the shelf once, while in fashion retailing merchandise is often handled repeatedly. Consider for example pullovers presented on tables which have to be folded again every time a customer goes through the pile, the amount of work needed depends largely on customer behavior. Direct Product Profitability does not consider any type of actual customer interaction except for the cash desk. The detailed Multi Moment Recording used at C&A Germany measured the actual time spent helping customers with different articles. Another significant difference is in the manner in which the use of space is calculated and costs allocated. In Direct Product Profitability, the amount of space used is the designated shelfspace for that stock keeping unit (SKU). Supermarket planogram software plans and keeps track of the exact amount of space used per article. In the unit cost research the amount of space used was a fairly rough estimate based on sample figures. As mentioned previously, the unit cost research allocated some costs based on the amount of space used but not rent or depreciation of the building. In Direct Product Profitability the cost of space can be handled in two ways. The first allocates all costs of the selling of the products including rent and overhead. This can lead to long discussions over the value of square meters. The historical cost of construction might be in large part sunk, or the rent fixed, but that does not necessarily mean that the cost of space is low. The relevant cost of 18-6-09 65 the space is the profit contribution that will be lost when some other product is removed from the store to make room for a new product. The profit contribution from the other product, per square meter, is the opportunity cost for displaying something new (Nagle, 1987). One can also take into account that certain areas within a store can have different values. The location of one store might be much more valuable than the location of another store which doesn‟t have as much customer traffic coming by. A part of the store can even be more valuable for a certain article than another part of the store. A supermarket, for example, probably wouldn't sell a lot more potatoes if they were presented at the checkout, but might lose sales if candy was moved to the back of the store. The second system for presenting DPP is simpler than the first and less „political‟ because it does not allocate costs associated with the square meters such as rent, energy, cleaning, local taxes, decoration, etc. This is the direct contribution per meter or DPP/sq. meter analysis. It is easier to understand and implement than a fully integrated calculation. The Assortment Model, which is the subject of the next section, discusses factors that influence contribution per square meter. 4.5 The Assortment Model In this paragraph, an idea developed by Kevin Corcoran (1994) for the C&A International Economic Department (IED) called the assortment model is discussed. The assortment model in contradiction to the unit cost research does not focus on selling, general and administrative costs. It is a model of revenue. It considers the factors that make up gross margin dollars per square meter. The IED suggested that it should be possible to authorize local executives to take decisions thatbecause of local knowledge-they are best positioned to take regarding different components of profitability including gross margin percentage of sales. This idea of decentralizing authority to store level is referred to as micro-marketing. The assortment model is based on the premise that although maximizing sales per square meter is widely held to be attractive, the more important objective is to maximize gross margin per square meter. In other words Gross Margin Return On Space (GMROS) also known as contribution per square meter. 18-6-09 66 The assortment model is a refinement of the link between GMROS and Gross Margin return on Investment (GMROI) described by Lusch in the Trinity model: GMROS or Gross margin dollars = Gross Margin dollars X Inventory Investment Square meters Inventory Investment Square meters In the assortment model the two components of the Trinity model mentioned above are each split into two sub-components. GMROI is based on the sub-components: quantity stock turn and gross margin percentage, while Inventory Intensity is divided into selling price and density. Density is the number of units per square meter. The resulting formula is: GMROS = GMROI X Inventory Intensity = GMROI X Inventory Intensity Gross margin dollars = Stock turn X Gross margin X Average Selling X Density of units per square meter in Quantity percentage Price per square meter Gross margin per square meter is thus broken down into four components: -Stock turn in quantity -Gross margin percentage -Average selling price -Density (stock per square meter) The first three components have already been discussed in this paper. The Assortment Model may be unique in the use of the concept of density. In clothing retail where items are constantly changing and styles are split up and presented by size (as described in figure 6), there is no space specifically assigned to one style. In the Assortment model calculation, density is used to bring space into the equation. In essence, using density to estimate space used, is a very simple idea. The total amount of space used is a result of the amount of room each single product typically occupies when presented on the selling floor ie. the density. 18-6-09 67 4.5.1 Density of units per square meter Density depends on two factors, the type of article and the way the article is presented. A pair of socks will in general take up less space than a coat. When fixtures are used to hang two sets of blouses on top of one another, the density will be higher than when the blouses are presented on a single level. Space and density are closely related. If more space is allocated to a product without increasing stock levels, density will decrease, and vice versa more stock on the same space entails a density increase. This is especially important for commodities such as trousers, where the number of pairs on a given rack can vary considerably. The selling price of the article presented does not directly influence density. It is not always clear what is meant by space used in a store. Stores do not only consist of selling space. Space is also used by such things as cash desks, walkways, pillars, fitting rooms, display windows, play areas, customer service areas, and display points (mannequins). It is important when defining density to carefully specify which space in a store is taken into consideration and which part is not. There are two ways of thinking about density. In the concrete case density is simply the quantity of stock of a commodity on the sales floor divided by the square meters it occupies. Thus, if a retailer is short of space, he can change the method of presentation. One could think of folding the item rather than hanging, or by double hanging instead of single hanging. Corcoran refers to the statistic used to measure this as store density. The Assortment Model calculation becomes a bit confusing when used in practice because of the manner in which stock turn is calculated. The inventory figures used to calculate stock turn generally come from retail information systems that include all allocated stock-including stock in the store reserve and out of house reserves such as in the distribution center. The result is that the density figure used for the calculation also has to include all allocated stock. This is referred to as total density. The figure below (which is essentially the same as Figure 5) shows the stock of two stores with the an equal amount of space and the same presentation methods. The stock turn and store density for both stores is equal, but store A has a higher GMROS. The total density which is used for the Assortment Model calculation is higher for store A than B. 18-6-09 68 New goods Distribution Center Higher "total density" DC Store A Store B Equal store density: same stock turn, lower sales per sq. meter Sales In summary: Store density = Stock of Y actually on the sales floor of commodity Y Square meters of space designated to Y Total density = Stock of Y on sales floor + all other allocated stock of Y of commodity Y Square meters of space designated to Y The alternative pricing methodology, which is explained in the next chapter does not make explicit use of density at all. The fundamental idea however of calculating space used in the store based on the number of units in stock was inspired by the concept of density. The Assortment Model is a useful tool for illustrating trade-offs that can be made in retailing. Recall the basic relationship: GMROS= Stock turn x Gross margin % x Price x Density The basic idea behind discounting (Chapter 1), low margins compensated by high stock turns is obvious. A high density and low margin is the rationale behind the strategy, "stack it high and sell it low". The different stock turn targets at C&A for low, medium and high (L,M,H)priced trousers can also be explained using the formula above. Expensive boutiques with high prices can afford to have a low density of articles. The fact that the Assortment Model is only concerned with revenue, limits it's usefulness to some extent. When considering how to present shoes for example, one might decide, based on the 18-6-09 69 Assortment Model, to present only one shoe of every pair on the showroom floor to double the density. The amount of space used to present a number of styles would be cut in half, and if sales were not effected, GMROS would be twice as high. The extra work required to help customers and the probable positive ramifications for shrinkage would however not be reflected. One trade off that Corcoran failed to consider is that between gross margin percentage and price. When I was working for C&A I suggested selling high priced shoes at a lower gross margin percentage than low priced shoes. The response I usually received was, "The C&A cost structure mandates a margin of at least X percent to make a profit." Here again the limitation of the Assortment Model is that it does not include Selling, General and Administrative costs and does not provide an alternative for planning based on a store wide average gross margin percentage. That is the aim of the next chapter. 18-6-09 70 Chapter 5 Meters, the Alpha and Omega In this chapter an alternative pricing and planning methodology will be described. The main difference between it and traditional retail planning is that revenue and costs are no longer measured as a percentage of sales. They are related to selling space or square meters instead. Meters, the Alpha and the Omega, refers to the idea of using meters as the basis from to the beginning of planning at company level down to the end decisions concerning individual styles. The first part of this chapter will explain this idea in more detail. Once again a comparison will be made to product costing in manufacturing. The following section gives a global outline of the planning process for a company using this method. This is followed by business unit planning, in which the groundwork for actual pricing decisions is laid. The manner in which styles can be evaluated is explained. This is followed by a proposal for a new pricing strategy, which is illustrated with practical examples. Finally foreseeable strengths and weaknesses of the alternative pricing methodology are discussed. 5.1 Changing the indirect cost application base The historical background of product costing and the reasons for its use in manufacturing and the lack of it in retailing were discussed in chapter 4. Three figures are presented below which represent an example of product costing in manufacturing, in traditional retailing and an alternative. Figure 10 is an example of manufacturing product costing. The cost of the product is made up of two basic components, direct and indirect costs. The direct costs are the costs of the materials used to make the product and the direct labor involved. The second component, indirect costs is calculated independently of the direct costs. In this example the indirect costs are based on the machine hours needed for the manufacturing process. Given the following information for a certain product and an overhead rate of DM 80 an hour Actual direct materials cost DM 700 Actual direct labor cost DM 280 Actual machine hours 8 18-6-09 71 The calculated indirect manufacturing product costs would be: 8 actual machine hours x the budgeted rate of DM 80 = DM 640. The total direct costs would be DM 700 material + labor DM 280 = DM 980. Thus the resulting product cost is DM 640 + DM 980 = DM 1 620. Figure 9 Product Costing Overview Example (Horngren & Foster, 1991) Manufacturing Indirect cost pool Factory Overhead Indirect Cost application base Machine Hours Product = Cost Object Indirect Manufacturing Product Costs Direct Manufacturing Product Costs Direct Product Costs Direct Materials Direct Labor The costing overview shown above can be used as a framework to analyze the retail process. In traditional retail planning the direct costs are the costs of goods sold (COGS). The indirect product selling costs are all the other costs, the Selling, General and Administrative (SGA) expenses. In retailing the direct and indirect costs are not calculated independently of one another. It can be argued that the direct costs are used as a sort of cost application base for the indirect costs. Retail: SGA (indirect cost)= COGS(direct cost) x cost multiplier (overhead application rate) On the following page the product costing overview is given for conventional retailing as well as for an alternative that I think is feasible. The alternative shown in the figure below can be used as a basis for retail price setting when using cost orientated pricing. The selling price is still made up of Cost of Goods Sold plus a margin to cover Selling, General and Administrative (SGA) costs, but the SGA costs are defined differently: They are calculated independently of the COGS, the amount paid by the retailer for the merchandise. 18-6-09 72 Figure 10: Product costing overviews for retail Traditional Retailing Indirect cost pool Selling General and Administrative Costs Alternative Store and central overheads Indirect Cost application base Cost of Goods Sold Square Meters Product Cost Object Indirect Product Selling Costs Cost of Goods Sold Indirect Product Selling Costs Direct Product (Selling) Costs Direct Product Costs COGS COGS Direct Labor The alternative method of retail price setting involves all the three main resources involved in retailing: inventory (COGS), labor and space. The indirect product costs are a function of square meters used. The amount of space an article takes up on the floor and the amount of time it takes before it sold are the determinants of space used. Alternative retail: Indirect cost ≠ COGS(direct cost) x cost multiplier (overhead application rate) Indirect cost = Space x time The direct labor costs could be calculated using methods similar to the Multi Moment Recording described in Chapter 3. Direct labor is not included in the next section in order to keep it simple. 5.2 Company planning without an average gross margin percentage target 5.2.1 Company level planning In normal retail planning processes, retailers ask themselves three fundamental questions. "What will my sales be?", "What will my gross margin percentage be?" and "What will my costs be?" (See appendix for example of C&A D (Germany) planning) 18-6-09 73 Under the alternative system the actual level of sales and the average gross margin percentage are irrelevant. The two important questions are, "What will my gross margin be in absolute terms?" and "What will my costs be?" To evaluate and plan operations the company would approach at the sections of the store from the viewpoint of space occupied. For example if the company was split into 3 different divisions, the global merchandise planning might look like this: Division Ladies Mens Children Total Meters occupied 30 20 10 60 Gross margin in DM 3 000 2 200 900 6 100 Gross margin DM sq. meter 100 110 90 102 To increase margins it might be sensible in this case to increase the area allocated to the mens division. As it stands Selling, General and Administrative costs would have to be under 102 DM/sq. meter for the company to make a profit. 5.2.2 Business unit level planning A business unit, for example the trouser commodity, would base its own plans on the number of meters allocated to it and its gross margin target (or history). Depending on the expected realized margin per unit(pair of trousers) the planned number of units sold could go up or down. Given 10 square meters and a gross margin target of 1 000 DM, the planning for the trouser commodity might look like this: Margin per pair: 5 DM Planned sales: 200 pairs (1 000 DM/5 DM) If the store density is 10 pairs per square meter, then a stock of 100 pairs (10 pairs per meter x 10 meters) could be displayed in the store. The stock turn would then roughly be 2 turns per year (planned sales 200 pairs/100pairs stock). Each business unit should have the authority to decide for itself what the margin per unit sold should be depending on a multitude of factors. These include the price level of the competition, 18-6-09 74 market positioning and the price elasticity that can be different for every article within a store (Van der Kind, 1996). 5.2.3 Style contribution In this section the implications of the alternative pricing methodology are discussed on the level of individual items. First the difference between a style and stock keeping unit (SKU) is explained. Style is the level at which decisions are made in fashion retailing. Retailers measure the range of merchandise assortment in stock keeping units (SKUs). Each specific product item in a merchandise group, that is distinguishable from all other items in the store represents one SKU (Ghosh, 1994). For example, a blue, button-down Oxford, extra long sleeve, Westbury shirt in a specific size constitutes an SKU. A store can stock quite a large number of SKUs within a product category because of the number of factors which can be distinguished. This is easily evident in the example shown below. Trouser Category Stock 3 Classifications or brands of trousers in 10 styles in 3 colors in 8 sizes Total SKUs 3 x 10 x3 x8 720 Compare the number of SKU's for the men's trouser department shown above to a typical Aldi supermarket that has less than 700 SKU's in total. The large number of SKU's in clothing is however not the only reason that calculations are not relevant for this level of detail. Decisions are not typically made at this level. In general retailers do not take a certain style and markdown the articles in sizes 44-48 or have different markups depending on the color. Generally, buyers and merchandisers will consider the performance of individual styles in their part of the assortment once a week. A style's contribution would be calculated by comparing the gross margin amount it generates minus direct labor with the meters it uses in the store. Style contribution is basically gross margin return on space (GMROS) minus costs of handling. If 10 pairs of bell-bottom trousers are sold at a margin of 5 DM per pair, the gross margin created is 50 DM. The amount of space used can be calculated using the average stock of the style (here bell- 18-6-09 75 bottoms), the total stock of the commodity and the amount of space allocated to the total commodity. Space used by a single style: Bell-bottoms stock 20 pairs Total trouser stock 110 pairs The resulting GMROS for the week is then gross margin/space used = 50 DM/1,8m² = 27,8 DM/m². Direct labor can be included in the calculation by subtracting the cost of direct labor from the margin per unit. 5.3 Using the alternative method in practice In this section practical examples will be discussed of the way articles can be priced and the different conclusions one might draw depending on the system used to evaluate sales performance. The selling prices are not determined by multiplying the buying price by a certain factor. An absolute amount is added instead to cover the actual costs of selling. The result is that within groups of similar goods the gross margin percentage of articles decreases as the selling price increases. Figure 11 A Possible Pricing Strategy This strategy is possible using existing retail systems. Problems would arise however if the store started selling more expensive goods Gross margin % X 10 m² (total space allocated for trousers) = 1,8 m² Gross margin % in relation to price (equal stock turns) than expected. To reach its average gross margin percentage goal, it would have to sell more low priced goods or increase the margins on the high priced goods. Using the Low Medium Average price High variable margin strategy illustrated in Figure 10 would make sales figures for articles or statistics on sales per meter meaningless. 18-6-09 76 Figure 11: Sales (per square meter) no longer meaningful This figure shows the result of selling Sales per square meter (equal GMROS) different articles at different margins. The income per square meter and thus the profitability for each article is equal, but the sales per square meter tell another Margin COGS Dollars per sq. meter (misleading) story. The thinking behind this is that the COGS (cost of goods sold) is not income for the Low Medium Average price High retailer but for his suppliers. The concept is easy to grasp for supermarket retailers who often pay their suppliers after they have actually sold the products involved to the end consumer. The margin is for the supermarket, the COGS component goes to the supplier. A supermarket that uses this basic strategy is the Aldi. Dieter Brandes, who used to be an executive director of ALDI Nord in Germany, asserts that Aldi sells a Margaux red wine there for 17,98 DM, which competitors using a traditional cost multiplier (based on an average gross margin) sell for at least 35,- DM. A comparison is made below between selling water and wine. A wine bottle is comparable to a water bottle in shape and size. The physical handling is not more difficult. For the example the handling costs are estimated to be 0,20 DM per bottle, which is achieved by moving the bottles on pallets and in boxes. The sales figures are based on the fact that "The shelf space allocated to an article at Aldi is generally equal to one week's demand." (Brandes, 1998) Three styles of trousers are also considered. Whereas at C&A Germany a trade off was made between selling price and stock turn (low prices needed higher turns) here a trade off is made between selling price and gross margin percentage. The direct labor is estimated to be 3,- DM per pair which is subtracted from the margin when calculating the contribution. 18-6-09 77 Example Trouser commodity 300 sq. meters, 15 sq. meters for the water, 3 sq. meters for the wine, direct labor 3,- per pair of trousers , bottle 0,20 DM Buying prices: Trouser L 24 DM, Trouser M 47 DM, trouser H 70 DM, bottle of water 0,50 DM, bottle of wine 17,10 DM Traditional style report Style Trouser L Bell Bottom Trouser M Corduroy Trouser H Grey wool Bottle of water Bottle of wine Gross Margin % 40 % 27% 20% 33% 5% Selling Price 40 DM 64 DM 88 DM 0,75 DM 18 DM Stock 3 000 3 000 3 000 3 000 600 Sales/wk 500 500 500 3 000 600 Alternative style contribution report for the same articles: Style Trouser L Bell Bottom Trouser M Corduroy Trouser H Grey wool Bottle of water Bottle of wine* Margin per Unit 16 DM 17 DM 18 DM 0,38 DM 0,90 DM Selling Price 40 DM 64 DM 88 DM 0,75 DM 17,98 DM Stock 3 000 3 000 3 000 3 000 600 Sales/wk 500 500 500 3 000 600 Style Contribution 65 DM/m² 70 DM/m² 75 DM/m² 10 DM/m² 140 DM/m² Both reports show the selling price, the stock and sales in units. The alternative style report also shows the amount received by the retailer each time a unit is sold (margin per unit). The style contribution is an added aspect. Given the top style report executives at C&A would generally be unhappy with the high priced grey wool trousers, because the margin (20%) is considered too low to cover selling, general and administrative costs. The style would not be repeated or if it was then the price would set at well over 100,- DM. 18-6-09 78 Any article priced with a 5% margin would normally be considered a give away, but in this case wine is the most profitable product. This is due to its low handling costs, high density, (relatively) high price and stock turn. 5.4 Strengths and weaknesses of the alternative method The alternative pricing methodology should help in the discussions on mark-downs, mark-ups, assortment planning, etc. I believe that by highlighting the difference between selling products of high and low value it can allow a retail to offer customers interesting prices without making a loss. One of the main advantages is that it allows flexibility in planning. One business unit can change its prices without another business unit having to change theirs in order to compensate the difference in the company's gross margin percentage. A disadvantage is that it is more complex than existing systems and maybe hard to grasp for many people. A precondition is that the square meters are allocated to the different product groups and are kept up to date in the information systems. Planning will also be difficult due to the circular cause and effect of the prices and costs. Lowering prices may increase sales thereby lowering labor costs per unit. In a store price has influence on costs, which can influence price. In a factory the production processes regardless of the selling price determines the time it takes to produce something. The product costs in manufacturing will not be influenced by the selling price. In everything you do in retailing you have to keep an eye on the business environment. You cannot just follow a system blindly. Most importantly a system or computer can tell you what you have sold and maybe how profitably, but it cannot tell you could have sold by having other merchandise in the store. 18-6-09 79 Chapter 6 Conclusion In the preceding chapters the development of the average gross margin method for pricing has been discussed. It was a useful tool to simplify retail operations and accounting before the widespread use of automation. Even today however the vast majority of retailers still set a single company wide average gross margin target which implies a direct link between cost of goods sold and selling, general and administrative costs. The literature on factors that influence retail performance is focussed on stock turn and margins. The factor inventory intensity is seldom considered. Assortment breakdowns into classifications at most non-food retailers do not distinguish between high and low priced articles within a commodity. In principle the value of the article being sold, is just as important a factor for profitability as its margin or stockturn. Few non-food retailers capitalize on the possible tradeoffs between value and other factors discussed in Chapter 4. Retailers involved in the food sector have been moving to towards systems which do reflect these nuances for over a decade. With a relatively low number of SKU's, simple labor processes and detailed space allocation systems, companies like Aldi estimate the actual costs for every product. in their assortment using absolute numbers not percentages. They have used this in selling everything from peas to PC's. In non-food some special issues must be addressed before more flexible pricing is possible as well. These are the costs of labor in the stores and the use of space by individual styles. Through systems like Multi Moment Recording it is possible to allocate store labor costs to individual product classifications. This helps distinguish between products where an outsider would at first glance consider the costs similar. For example the store staff labor costs for ladies shoes is more than twice as high as for men's shoes. Because (German) ladies tend to try on a number of different pairs and do not put them back in the racks. The second barrier to be overcome is the use of space. The simple idea of using the number of units in stock to estimate the amount of space to allocate to a style and/or classification is not 18-6-09 80 widely documented in retail literature. Within a complex environment of constantly changing styles and presentation, this system is relatively simple and easy to understand. On the whole the fundamental changes in planning and reporting necessary to implement the system should not be underestimated. On the on hand the (IT) systems and operating procedures would need to be changed, on the other hand changing the views held by the people working in a organization is never simple. The felling of loss of entrepreneurial control which top management might feel should not be overlooked. Further research could include a more detailed analysis of the exact calculations used by different companies such as the Aldi and Wal-Mart which both sell a considerable amount of clothing. The best way of finding out whether the ideas expressed in this paper are realistic is by putting them into practice. The proof of the pudding is in the eating. Many unforeseen issues and interesting problems will undoubtedly surface, but that for that to happen it has to be tried first. 18-6-09 81 BIBLIOGRAPHY Arnold, J., Turley, S.,(1996), Accounting Management Decisions, 3rd edition, Harlow: Financial Times /Prentice-Hall, Inc. Borin & Farris, (1990), "An Empirical Comparison of Direct Product Profit and Existing Measures of SKU Productivity", in: Journal of Retailing, Volume 66, Number 3, Fall 1990 Brandes, D.,(1998), Konsequent einfach: Die Aldi-Erfolgsstory, Frankfurt/New York: Campus Verlag Brenninkmeijer, A.J.,(1996), Using Activity Based Costing to Model Supply Chain Decisions, The Hague: Konmar Superstores Broekman, T.,(1990), The Legend of Clemens & August, Printed in W.-Germany Buuron, P.M.A.,(1995), Retail Informatie Technologie, Concurreren met Informatie Technologie, Amsterdam: VU Uitgeverij Cooper R., Kaplan R.S, (1988)., “How Cost Accounting Distorts Product Costs”, in: Management Accounting, april 1988 Cooper R., Kaplan R.S.,(1988), “Measure the Costs Right: Make the Right Decisions”, in: Harvard Business Review, september-october 1988 Corcoran, K.R.,(1994), Managing Assortment Profitability: Tools for Selling and Merchandise Management, Amsterdam: IED Cox, R., Brittain, P.,(1996), Retail Management, 3rd edition, London: M&E/Pitman Drucker, P.F.,(1964), Managing for Results, New York: HarperCollins Drucker, P.F.,(1980), Managing in Turbulent Times, New York: HarperBusiness Eglau, H.O.,(1972), Die Kasse muß stimmen: So hatten sie Erfolg im Handel, Von der Kleiderdynastie Brenninkmeyer über die Discountbrüder Albrecht bis zur Sexversenderin Beate Uhse, Dusseldorf: Econ Verlag Ghosh, A.,(1994), Retail Management, second edition, Fort Worth: The Dryden Press Goor van A.R., Ploos van Amstel M.J., Ploos van Amstel W.,(1989), Fysieke distributie, denken in toegevoegde waarde , Houten: Stenfert Kroese, Leiden/Educatieve Partners Horngren, C.T., Foster, G.,(1991), Cost Accounting: A Managerial Emphasis, 7th Edition, Eaglewood Cliffs: Prentice-Hall, Inc. 18-6-09 82 Janse de Jong, V., De With, E., (1999), "Direct Product Profitability: A More True Margin for Decision Making", in:Handboek Management Accounting, Alphen aan den Rijn: Samson Documentaire Uitgaven Kind, R.P. van der,(1996), “Retailmarketing”, Houten, Stenfert Kroese Lusch, R.F., Dunne P., Gebhardt, R.,(1993), Retail Marketing, second edition, Cincinnati:South-Western Publishing Co. Management Horizons,(1992),Europe's Leading Specialist Fashion Retailers, London: Management Horizons Ltd Miellet, R.,(1994), Honderd jaar Grootwinkelbedrijf in Nederland, Zwolle: Uitgeverij Catena Nagle, T.T.,(1987), The Strategy and Tactics of Pricing:: A Guide to Profitable Decision Making, Englewood Cliffs, Prentice Hall Nooteboom, B.,(1981), A New Theory of Retailing Costs, The Hague: Research Institute for small and medium -sized business in the Netherlands Nooteboom, B.,(1982), A Model of Retail Margin, The Hague: Research Institute for small and medium sized business in the Netherlands Ortega, R.,(1998), In Sam We Trust:: the untold story of Sam Walton and how Wal-Mart is devouring the world, New York: Random House, Inc. Osté-Ruizendaal, J.C.,(1994), Direct Product Profitability en de commerciele relatie tussen de levensmiddelhandel en industrie, Nijmegen: Mediaset Uitgeverij bv Pasdermadjian, H., (1950), Management Research in Retailing,, London: Newman Books Ploos van Amstel, M.J.,(1985), De relatie tussen produktkarakteristieken en de inrichting van het magazijn/DC, in: Praktijk boek Magazijnen/Distributiecentra, onder redactie van: J.P. Duijker, Dr. M.B.M. de Koster, Prof.jhr.drs. M.J. Ploos van Amstel, Kluwer Bedrijfswetenschappen, Afl. 3 oktober 1995 Porter, M. E.,(1985), Competitive Advantage: Creating and Sustaining Superior Performance, New York: The Free Press Stern, L.W., El-Ansary, A.I.,(1992), Marketing Channels, 4th edition, Englewood Cliffs: Prentice-Hall, Inc. Surrey Beheermaatschappij bv, (1973), De Eerste Drie Generaties van C&A, Amsterdam Thurik, A.R., Koerts, J.,(1983), Further study on the use of retail floorspace and its efficiency, The Hague: Research Institute for small and medium -sized business in the Netherlands 18-6-09 83 Van Abswoude, Q.J.C.,(1997), "Direct Product Profitability in de detailhandel", in:Handboek Management Accounting, Alphen aan den Rijn: Samson Documentaire Uitgaven Walton, S.,(1992), Made in America, My Story, New York: Doubleday Wileman, A., Jary, M., (1997), Retail Power Plays: From Trading to Brand Leadership, London: Macmillan Press Limited Willard Bishop Consulting, Ltd. (1997), Introducing CPCM Category Profit & Cost Model: "The Practical Activity-based Costing Tool for Category Management", Barrington, IL: WBC Zimmerman, R.M., Kaufman R.M., Finerty G.S., Egan J.O.,(1990), Retail Accounting and Financial Control, Fifth Edition, New York: John Wiley & Sons 18-6-09 84 APPENDIX -Unit cost research Multi Moment Recording form -Space used form -Planning based on margin and sales as basis for business unit planning -Excerpt from SAP Retail manual