Copyright MCB UP Limited (MCB) 1998Robert F. Lusch: Michael F. Price College of Business Administration, University of Oklahoma, Norman, Oklahoma, USA
Stephen L. Vargo: California Polytechnic State University, California, USA
Introduction
Alternate distribution channels for supplying the business or commercial customer in the US have become more common over the last two decades. Increasingly office supplies and equipment are being distributed to the business customer through office supply superstores such as Office Depot, Office Max, and Staples which sell approximately 80 per cent of their merchandise to business/commercial accounts. Building materials and related products are being purchased by general and specialty contractors and other business/commercial accounts from warehouse home centers such as Home Depot, Home Base, Home Quarters, and Builders Square which sell approximately 20-30 per cent of their volume to these customers. And everything from food supplies, to office supplies and industrial supplies and equipment is finding its way to the business customer via membership warehouse clubs such as Sams and Price/Costco which sell about 50-70 per cent of their merchandise to business/commercial accounts.
Importantly, within the last five years membership warehouse clubs from the USA have entered Canada, Mexico, South America, Europe, and Asia (Chanil, 1994; Ferguson, 1997; Thornton, 1994). Furthermore the US office supply superstore chains have also entered these markets. Coupled with this expansion is the power of the Internet and the ability to market globally. For instance Wal*Mart has established Sam's Club Exports which offers a wide assortment of consumer merchandise available from one source in name brand and private label non-perishable food, and hard-line products. This division of Wal*Mart is in the business to sell small- and medium-sized businesses at a lower cost structure than other types of distribution, allowing for cost savings and thus greater value. Orders must consist of a single item in full container quantities from one manufacturer. If full container quantities are not preferred, a $10,000 USD minimum per item is required.
Because these new forms of power retailing are restructuring distribution channels and because of the breadth of their impact, a research study was conducted to help answer the question of why businesses are purchasing from multiplex retailers. In this study we compare traditional wholesale sources of supply with these new multiplex retailer supply sources. To shed light on our fundamental question we use a "total value of purchasing" model as a theoretical framework. Implicit in this model is the assumption that businesses select suppliers in order to maximize their perceived total value of purchasing and suppliers offer them services to help them accomplish this goal. This model is not new; rather, it is distilled from a number of overlapping models from the organizational buyer behavior and marketing channels literature. What is unique to this study, unlike previous tests of the related models, is the use of data representing actual perceptions and behavioral intentions. Additionally, the study represents the only investigation of buyer choice involving the quickly growing alternative channel for wholesale distribution of multiplex retailers. The theoretical rationale for this model is presented and then it is used to assess the determinants of patronizing traditional wholesale supply sources versus multiplex retailer supply sources. Before we proceed, however, a brief review of multiplex retailing is provided because this largely US phenomenon has global potential for explosive growth which would disrupt many traditional marketing channels around the world.
Multiplex retailing
A multiplex retailer is one that targets both the traditional household market as well as the business or commercial market. Thus with one distribution format these firms attempt to appeal to two distinct market segments by performing both wholesaling and retailing functions. Although it is true that multiplex retailing is not new; for example, hardware stores have historically sold to contractors and the local grocery store would frequently sell to restaurants short on supplies; multiplex retailing is gaining momentum because of the number and variety of power retailers pursuing the multiplex retail strategy (Lusch and Zizzo, 1995).
Multiplex retailing coupled with power retailing is sending a clear signal to all channel members in virtually all industries: stop and consider the future of this distribution format. This is because power retailers have such a high degree of relative power that suppliers (either manufacturers or wholesaler-distributors) are unable to counteract this power without a substantial investment of time, effort, and financial resources. In effect, the retailer becomes the "captain" of the marketing channel. For some companies, power retailing is a strategic advantage that provides prolonged competitive strength in both the marketplace for consumer and business purchasing.
To gain a better insight into the history and growth of multiplex power retailing we briefly review the two lines of trade studied in this research: warehouse home centers (pioneered in 1979) and office supply superstores (pioneered in 1986).
Warehouse home centers
In 1979, Bernard Marcus and Arthur Blank opened in Atlanta, Georgia three Home Depot stores of approximately 65,000 square feet in size and handling 18,000 SKUs. In 1981 Home Depot became a publicly held company and two years later operated 19 stores. That year, 1983, was a key year for the industry because it witnessed the spread of this concept to other parts of the country: Builders Square in Texas; HomeClub in California; and Home Quarters Warehouse in Virginia. Currently, about a dozen companies operate the "big box" warehouse home center format stores across the country. One of the latecomers to the warehouse retailing format has been Lowe's. In the 1970s, Lowe's was a pioneer in the do-it-yourself market with a chain of small, showroom-type stores, and, in fact, was the industry leader in terms of sales. In the late 1980s, however, sales gains began to slow, and by 1990, Home Depot had overtaken Lowe's as the largest home improvement chain. Finally, in the early 1990s, Lowe's announced a major repositioning strategy to enter the warehouse home center industry. Today, warehouse home centers operate in most major metropolitan areas and have begun entering smaller markets as well. Warehouses usually average 100,000 square feet, primarily selling space, with an additional 10,000 to 15,000 square feet outside devoted to lawn and garden. These warehouses typically stock 30,000 SKUs but some of the larger stores stock up to 50,000 SKUs. Indeed warehouse home centers, serving both the professional customer and household customer, have become a formidable competitive force in US distribution. Meanwhile Home Depot has entered Canada and is considering future international expansion.
Office supply superstores
The office supply superstore traces its roots to both membership warehouse clubs and warehouse home centers. In 1986, two companies independently opened the first stores in this format. Staples opened in Massachusetts in May, and Office Depot launched its first store in Florida in October. Staples' co-founders Leo Kahn and Thomas Stemberg were from supermarket retailing backgrounds and had observed the success membership warehouse clubs were having with office supplies. The three founding partners of Office Depot wanted to apply Home Depot's techniques - large, well-stocked, no-frills stores - to the office supply industry. In fact, Office Depot copied so well, right down to the name, that they had to state there was no affiliation with Home Depot for six months and give that company $25,000 worth of office supplies to keep Home Depot's legal department at bay. These companies were dedicated to shortening the channel of distribution for office supplies. Staples and Office Depot, by purchasing direct from the manufacturer, were able to sell office supplies and products for 20 to 70 percent less than other retailers.
Since its inception 12 years ago, the office supply superstore industry has already experienced the rapid growth and consolidation phases of its life cycle. Many similar companies were started and expanded rapidly before their numbers peaked in 1989 at 19 chains. Consolidation was just as rapid and fierce over the next few years, until today the industry is dominated by three companies: Office Depot, Office Max, and Staples. Today, most of these stores are in the 18,000 to 25,000 square foot range and stock 5,000 to 6,000 SKUs covering a broad range of office supplies and equipment. This type of method for distributing office supplies and products can also be expected to experience global growth. For instance, Office Depot went international in 1992 when it acquired the Great Canadian Office Supplies Warehouse chain and converted the five stores into Office Depots and began opening stores throughout Canada. Next in 1993, an international licensing agreement was signed which resulted in Office Depot stores in Columbia and Israel. Shortly after that, in 1995, a joint venture agreement with Grupo Gigante led to stores opening in Guadalajara, Mexico. Also in this year the first store in Poland was opened and in 1996 two stores in France opened under a joint venture with Carrefour. In 1997 two stores were opened in Thailand in cooperation with Central Department Stores. A joint venture with Daiichi Corporation should result in the first Office Depot in Japan in 1998.
The total value of purchasing model
In a search for why these firms may have succeeded, we sought an explanation for why businesses, which have traditionally purchased office supplies and hardware supplies from wholesale-distributors, may be purchasing from office supply superstores and warehouse home centers. In this search we were able to distill a model of the total value of purchasing. This model considers the total perceived value of purchasing and how business buyers attempt to minimize cost and maximize value of services obtained.
Whether purchasing components to be used in production, or capital goods and consumables to be used in the production process, commercial purchases are made with the intention of profit realization. Defined in terms of maintaining costs at a level below revenue, profit is a rather straight-forward concept. However, the identification of the relevant "costs" of a particular product is not always a simple matter. For example, two alternative components to a production process may have the same purchase price but different overall costs due to quality differences; or they may have the same purchase price and quality but different overall costs due to differences in supplier reliability in delivery, credit terms, technical support, problem recovery, etc. That is, total product costs comprise not only purchase price, but also costs associated with the full range of supplier services provided (or not provided). It is from this total value (a function of perceived total costs) perspective that we explore how business firms make supplier choices between a traditional wholesale-distributor and the new emerging multiplex retailers - specifically, office supply super-centers and warehouse home centers.
This total value of purchasing model is not entirely new. However, the present study differs from previous studies in several important regards. First, previous studies have tended to focus on suppliers which manufactured and supplied a single (or limited range of) product(s). Consequently, product quality and other supplier services were inextricably intertwined. In the present study, however, both wholesale distributors and multiplex retailers offer a wide variety and often the same or similar products, though possibly different assortments (a service issue). Second, since most previous studies have focused on different suppliers supplying specific products, product/supplier selection analyses have usually been limited to the comparison of attribute importance weights rather than actual supplier selection intentions. The present study offers a relatively unique opportunity to analyze supplier selection intentions as a function of the total value of the services purchased, under conditions in which "product quality" is relatively controlled.
Supplier analysis
Two general approaches to evaluating alternative suppliers are found in the organizational buying literature. The first is a normative approach concerned with the development of optimal models for comparative analysis of alternative products or suppliers. Related literature is usually conceptual but may be derived from or offer case study analysis for support. The second approach is more positively oriented and consists of empirical investigation of relative attribute importance employed in different buying situations. What is missing in this literature is the empirical investigation of actual perceptions of decision criteria in relation to intended behavior.
Normative models
Normative models are typically extensions of one of three supplier evaluation models that are purported to be commonly employed in industry (Giunipero and Brewer, 1993; Timmerman, 1986). These three models are the "categorical approach," the "weighted-point plan", and the "cost-ratio approach," and are summarized in Table I.
Several of the normative approaches (e.g. Gregory, 1986; Timmerman, 1987) simply recommend methods for standardization and the "objective" evaluation of criteria. Others propose more extensive modifications. For example, Thompson (1990) proposes a modified weighted-point model that replaces single-point supplier evaluations with ranges for each evaluative criterion. Graphical representation of these ranges can be used to compare potential suppliers.
Some normative models recommend multiple-step approaches to supplier selection (see Table I). These models (e.g. Giunipero and Brewer, 1993; Smytka and Clemens, 1993) typically involve subjective evaluation of relationships followed by more objective evaluation of preselected alternative suppliers. Ellram and Siferd (1993) (see also Ellram, 1994) group all of these approaches under the rubric "Total cost of ownership" (TOC), and Ellram (1995, pp. 12-13) distinguishes between two approaches to TOC: a "dollar-based" approach, based on actual costs; and "value-based" approaches, which "combine dollar/cost data with other performance data which are difficult to 'dollarize'." The trend in these prescriptive studies is clearly in the direction of the incorporation of performance, improved quality, relationship, and other service factors that affect total value, in addition to the more traditional and narrow concerns for selling price.
Positive analyses
This trend in purchasing management toward an emphasis on quality and service, in addition to selling price, is partially supported by empirical studies intended to assess evaluation criteria actually employed by organizational purchasers. However, relative attribute importance has been noted to vary across buying situations and product categories. For example, Lehmann and O'Shaughnessy (1974) found that relative importance weights assigned to attributes by purchasing agents varied considerably across product types. Reliability of delivery was ranked as important across all product types. However, price was more highly ranked for products that were routinely ordered and for products where there was interdepartmental disagreement about the evaluation on other criteria; service attributes such as technical support were more highly ranked for products which required training and behavioral change. In a follow-up study, Lehmann and O'Shaughnessy (1982) found economic criteria (e.g. price) to be most important for standard products of simple make-up, standard application, and low dollar value, while performance criteria were most important for products of complex make-up and/or novel application.
Doyle et al. (1979) found that delivery, price, and payment-terms were most important in straight-rebuy situations, while price, product performance, delivery, and guarantee were most important in first-time buying and modified-rebuy situations. However, they also found that post-purchase evaluation and search for potential suppliers was very limited in most industrial rebuy situations. Puto et al. (1985) found evidence of a "loyalty barrier" in industrial buying. Specifically, they found that in a modified-rebuy situation, the current supplier was more likely to be selected even when an alternative supplier offered less risk (e.g. a guarantee). That is, supplier loyalty seemed to mediate risk taking.
Several researchers have noted differential relative attribute importance as a function of product purpose. For example, Kauffman (1994) found evidence that for products purchased for use in the "production process" (components), all individual attributes ("physical", "non-physical", "price", and "distribution") were important, while for "capital equipment" products and "administrative" products physical and distribution attributes were more important. The implication seems to be that while service, economic, and physical attributes are important for production products, service is more important for the other product classes.
Wilson (1994) in comparing a review of past studies concerning relative product and supplier attributes with a current survey of buying center members, suggests that there has been a general shift in relative attribute importance over the past 20 years. Specifically, she notes a shift away from price and delivery and toward "service" (service-call response time) and quality (durability). She (Wilson, 1994, p. 36) sees this shift stemming, in part, from globalization and increasing competitiveness and contends: "the relationship of the quality and service factors to total product cost is an important element in this equation. A purchased product's total cost is made up of initial price, various direct and indirect costs associated with product quality, and a similar array of costs associated with the service required to support acquisition and use of the product. It is not surprising that purchasers who strive to minimize total cost place greater emphasis on quality and service and less emphasis on price."
Most of the empirical studies reviewed have several common characteristics. First they typically involve the sampling of purchasing managers of relatively large firms, and less often include the selection criteria of internal users. Second, they are usually based on the selection of suppliers offering a single product or limited range of products, and therefore supplier analysis is inextricably intertwined with product analysis. Third, most are based on relative importance weights, but do not investigate the perceived performance attributes of those suppliers actually selected. Fourth, most rely on single-measure indices of the modeled attributes of total cost. Finally, with some exception (e.g. Doyle et al., 1979; Puto et al., 1985), most do not explore the role of supplier loyalty, buyer inertia, and buyer-seller relationship in the purchasing decision. Of particular interest may be the role that these variables may play in "overriding" other attribute weights.
Present study
Middlemen, either wholesalers or retailers, can attract customers by providing value-added services, such as directly lowering costs or providing other perceived cost-reduction services. By adopting this perspective we view all activities performed by middlemen as services (cf. Stern et al., 1996). That is, whether classified as a wholesale-distributor or a retailer, the functions of entities that supply products to other entities are, by definition, services. Services that are not performed by manufacturers or middlemen must be borne by the purchaser. Therefore fewer value-added services represents a potential cost to the purchaser because they must perform the service for themselves that middlemen do not perform.
From this perspective, total value can be seen as a combination of the price and the indirect perceived costs that the purchaser incurs due to inadequate or fewer value-added services being provided by the middleman. These value-added services, include providing credit, product acquisition services, and product use services. We call the components of this four component total-value model: price, credit services, acquisition services, and purchase-risk reduction services respectively. This classification is similar to Lehmann and O'Shaughnessy's (1982) "economic", "integrative", and "adaptive" criteria classification. However, unlike the Lehmann and O'Shaughnessy classification, we see product performance, or more appropriately the quality of the product assortment offered, in terms of a service that reduces purchasing risk (i.e. an adaptive criterion). This classification is also consistent with the generalized "price", "delivery", "service", and "quality" classification used in Wilson's (1994) study, with service and quality combined into "risk-reduction" services. Further, the classification is consistent with the classification of channel intermediary "functions" from the channels literature; specifically, credit (economic services), after-sale service (risk-reduction services), inventory, and physical distribution (Stern et al., 1996), with the latter two classes combined into product acquisition services. Comparison of these various classification schemes can be seen in Table II.
Figure 1 illustrates the relationships we have been discussing between price, credit services, product acquisition services, risk-reduction services, and total customer perceived value and patronage preferences. This model suggests that as prices are lower and value-added services are higher, the customer's total value of purchasing increases. Consequently this increased total value leads to higher patronage preferences. Alternatively stated, we hypothesize:
H1:Supplier patronage intentions toward wholesale-distributors compared to multiplex retailers are positively related to buyers' perceptions of wholesale-distributors' relative pricing advantage compared to the multiplex retailer.
H2:Supplier patronage intentions toward wholesale-distributors compared to multiplex retailers is positively related to buyers' perceptions of wholesale-distributors' relative advantage in providing credit services.
H3:Supplier patronage intentions toward wholesale-distributors compared to multiplex retailers is positively related to buyers' perceptions of wholesale-distributors' relative advantage in providing acquisition services.
H4:Supplier patronage intentions toward wholesale distributors compared to multiplex retailers is positively related to buyers' perceptions of wholesale distributors relative advantage in providing risk-reduction services.
These hypotheses are generally consistent with previous purchasing models, but differ in two significant regards. First, they are stated in terms of actual buyers' perceptions of how alternative suppliers perform on various attributes (i.e. price and perceived service provision) rather than how these attributes are weighted based on their importance. Second, they explicitly link these perceptions to actual behavioral intentions rather than leaving the link implied. That is, while most of the previously cited studies assumed some form of a total value model, the related empirical research was limited to descriptive statistics concerning relative importance weights. The present study is designed to test explicitly the relationship between actual perceptions of price and service performance and actual buyer intentions.
The research method
A survey research design was used to gather the data necessary to test the proposed hypotheses. The research design consisted of:
- (1) identification of population to be surveyed and sampled;
- (2) questionnaire design and administration; and
- (3) scale validation and data analysis.
Population selection and sampling
The population was businesses that regularly purchased from wholesaler-distributors as well as from multiplex retailers. The multiplex retailers we were interested in studying were warehouse home centers and office supply superstores. Consequently, for each we identified, through exploratory research, a list of representative businesses that would be likely to purchase from the line of multiplex retailing under study. For warehouse home centers, businesses in 13 SIC codes representing the building and contractor trades and some lines of retail trade were selected. For office supply superstores, businesses in eight SIC codes representing service firms were chosen.
As regards the survey of geographic markets, two options were to survey business customers throughout the entire USA (in the SIC codes selected) or only in selected markets. Since warehouse home centers and office supply superstores currently operate in nearly every major metropolitan market in the country, a sample of medium and large cities was selected. The selection was eventually narrowed to six markets: two large markets and four medium-sized markets. The two large markets chosen were Dallas/Fort Worth, Texas, and Los Angeles, California. These two cities were, respectively, the 2nd and 7th largest wholesale markets in the USA. The four medium-sized cities selected were: Fresno, California; Grand Rapids, Michigan; Tulsa, Oklahoma; and Wichita, Kansas. All markets are among the largest 75 cities in the USA. In addition, multiplex retailers of both types were present in each market, and the level of competition ranged from moderate to high in each of these cities.
A total of 3,410 questionnaires were mailed during May 1994: 2,090 to potential customers of warehouse home centers and 1,320 to potential customers of office supply superstores. The smaller sample size for the office supply study was determined by the sponsor of the research who was interested in both types of multiplex retailers but felt that the warehouse home centers posed more of a competitive threat to a wider number of wholesaler-distributors. In the two larger cities we sent 525 questionnaires to potential customers of warehouse home centers and 330 to potential customers of office supply superstores. In the four smaller cities we sent 260 questionnaires to potential customers of warehouse home centers and 165 to potential customers of office supply superstores.
Questionnaire design and administration
The primary goal of the questionnaire, as it relates to this study, was to identify the services a wholesaler-distributor or multiplex retailer could offer businesses and how each of these alternative supply sources was rated on performance, as perceived by the business customer. Also collected were patronage data and a variety of demographic and descriptive data of the responding firms. Separate, but largely similar, questionnaires were developed for potential customers of the two survey groups. The questionnaires were refined and pretested on approximately 15 businesses in a major metropolitan area in the Southwestern, US.
The two questionnaires were mailed to 3,410 businesses as described earlier. Several things were done to increase the response rate: a personalized cover letter with a hand-written self-sticking note to the respondent urging their participation; a one dollar bill was included as a tangible symbol of appreciation for the respondent's time; first class mailing, with return postage-paid envelope enclosed; and a follow-up reminder postcard sent within five to seven days of the original mailing. The random sample of company names and addresses for businesses in the selected SIC codes were provided by American Business Lists of Omaha, Nebraska. Although this firm is a well-respected provider of mailing lists, about 6.7 per cent of the addresses were non-deliverable, due to businesses being terminated, sold, or relocated. Therefore, of the 3,181 questionnaires actually delivered, 1,211 were returned for a response rate of 38.1 per cent. Of those returned, 425 were usable for this study. The reduction in size was because we only analyzed those responses where the businesses purchased both at a wholesaler-distributor and a multiplex retailer (office supply superstore or warehouse home center) within the last year.
Scale validation
As suggested in the four hypotheses that were developed, we needed to measure buyers' perceptions of supplier performance on price, credit services, product-acquisition services, and risk-reduction services and patronage intentions toward suppliers. Table III shows the four service categories for the 21 specific service attributes used in the study.
For each of the 21 attributes in Table III, we asked the respondent to rate both the wholesale-distributor's and the multiplex retailer's performance on a scale ranging from "1" poor performance to a "9" outstanding performance. The relative measure of performance was thus computed as the score the wholesaler received minus the score the retailer received. It was these difference scores for which we computed reliability scores. These reliability statistics are reported in Table III. Peter et al. (1993) report that difference scores can be less reliable than their component parts. Consequently we use procedures these authors recommend to determine the true reliability of these difference scores. For the four measures used as independent variables (price, credit, acquisition, and risk) the reliability measures were between 0.68 and 0.88. According to Nunally (1967) reliabilities in the 0.50 to 0.70 range are adequate.
Patronage was measured with two questions. One question had the respondent indicate on a "1" to "5" scale the likelihood of their buying at the supply source (either wholesaler or multiplex retailer) in the future. The second question had the respondent reply on a "1" to "5" rating if they planned to buy more from the supply source in the future. These items were summed to obtain a total score and then the score the multiplex retailer received was subtracted from the score the wholesale-distributor received. Table IV shows the reliability for these difference scores. The reliability for the relative patronage preference for office supply superstores versus office supply wholesalers was 0.54 and that for warehouse home centers versus hardware wholesalers was 0.65. While these reliabilities are somewhat lower than ideal, they are also adequate according to the Nunally (1967) criteria. It should also be noted that the validity of the scales was not formally assessed although the items that comprise the scales came from field interviews and have face validity. Nonetheless, the formal validation of the scales is an area that future research needs to address.
Statistical model
To test the three hypotheses we developed the following regression equation.
RP = B[sub]o + B[sub]1P + B[sub]2C + B[sub]3A + B[sub]4R + e
Where:
B[sub]o , B[sub]1, B[sub]2, B[sub]3, B[sub]4 = regression parameters to be estimated.
RP = relative patronage preference (difference score) toward wholesaler versus multiplex retailer.
P = relative perceived performance (difference score) of wholesaler versus multiplex retailer on low price provision.
C = relative perceived performance (difference score) of wholesaler versus multiplex retailer on credit provision services.
A = relative perceived performance (difference score) of wholesaler versus multiplex retailer on product-acquisition services.
R = relative perceived performance (difference score) of wholesaler versus multiplex retailer on risk-reduction services.
Two regression equations are necessary, one for warehouse home centers and a second for office supply superstores. All four hypotheses predict positive values for the beta coefficients.
Results
The standardized regression parameters and related statistics for the two regression equations are presented in Table V. Both regression equations are statistically significant. The regression equation for the warehouse home centers had an F-value of 23.14 which was significant beyond the 0.0001 level and had an adjusted R-squared of 24.4 per cent. Similarly the office supply regression equation had an F-value of 19.17 which was statistically significant beyond the 0.0001 level and had an adjusted R-squared of 30.91 per cent. Overall these results suggest that the total value of purchasing model explains a significant amount of the variation in purchasing of businesses at traditional wholesale supply sources versus multiplex retailers.
In terms of the four hypotheses we find support in the warehouse home center sample for H1 and H4. As multiplex wholesalers outperform multiplex retailers on prices they attract more customers (support for H1) and as wholesalers outperform multiplex retailers on risk-reduction services they attract more customers (support for H4). On the other hand no support was obtained for the role of credit services (H2) and product-acquisition services (H3). This suggests that the business or commercial purchaser of hardware and building supplies is most driven by price and risk-reduction services. Especially in the building and remodeling trades, risk-reduction services are critical because errors in deliveries, stock-outs, unknowledgeable personnel, and so on, can create substantial costs to the builder or remodeler. This customer has a crew and equipment scheduled to do a job and a mistake made by the supplier (either wholesale-distributor or warehouse home center) can cost more than the product itself. A sense for the tradeoffs commercial customers are willing to make between price and risk reduction services can be obtained by examining the standardized regression parameters which are 0.17 for price and 0.31 for risk. If a warehouse home center performs poorly on risk reduction services than it can compensate for this only by a more than proportionate gain in price performance (i.e. lower prices). Alternatively wholesalers that perform well on risk-reduction services compared to warehouse home centers do not need to be as price competitive.
The results from the office supply superstore regression equation were even more strongly supportive of the total cost of purchasing model. In this situation all but the credit services (H2) were positively related to relative patronage preferences. Importantly all three of the standardized regression parameters had approximately the same value with the value on risk a bit higher. This suggests that a higher price can be traded off with improved product acquisition services and better risk reduction services. On the other hand if a multiplex retailer fails to do a comparable job to wholesale-distributors on product-acquisition services and/or risk-reduction services it will need to proportionately lower its prices to attract customers.
Discussion
A relatively simple total value model of supplier selection by business purchasers was able to explain from 24-31 per cent of patronage behavior. The model postulates that businesses purchasers decide between alternative suppliers based on the relative performance of these suppliers on price, credit services, product-acquisition services, and risk-reduction services. Its empirical support confirms previous research of organizational buyer behavior that has not previously been tested directly. As a supplier is able to offer lower prices and credit terms the purchaser's direct cost of purchase declines. At the same time as the supplier is able to offer more product-acquisition services and risk-reduction services the perceived indirect costs of purchasing decline for the purchaser and thus patronage toward this higher performing supplier rises.
Generally this model was supported, with the exception that credit services were not shown to be a significant predictor of supplier selection. Perhaps credit markets are highly efficient and business purchasers of office supplies and hardware recognize that they are paying the market rate for credit terms that are offered by suppliers. They may recognize that it is better to not take trade credit and instead finance purchases and inventory with other sources of working capital. The nonsupport for credit services in these two organizational buying settings is also consistent with Lehmann and O'Shaughnessy's (1982) conclusion that total cost is differentially determined as a function of product type and application.
The fact that this relatively simple model performed so well is a bit surprising when one compares it to retail patronage models of household purchasing. These models tend to be more complex and at the same time explain relatively little variation in retail patronage behavior (Darden and Lusch, 1983). Perhaps the purchasing of businesses is inherently more predictable or explainable than that of households. Of course, one possibility is that the models of retail patronage behavior of households may be misspecified. One might question if the total value of purchasing model might also be a good framework for explaining household retail patronage preferences. Do households choose between alternative retailers based on price, credit, product-acquisition services, and risk-reduction services? Can their behavior also be explained in terms of a simple total value model? This could be a fruitful area for future research. Competitive strategy implications can also be derived from the total value of purchasing model. Currently, wholesale-distributors feel threatened by multiplex retailers as these firms make more inroads into their markets (Lusch and Zizzo, 1995). These wholesalers often feel they cannot meet the prices of the multiplex power retailers. Our results suggest that they should not attempt to enter into price competition but instead concentrate on developing their product-acquisition services and risk-reduction services, thus enhancing the total value of the relationship (cf. Maltz and Ellram, 1997). These are services that traditional full-service wholesalers have excelled on and for which they can defend themselves against the power retailers. Offering these services increases the cost of business of the wholesaler and at the same time lowers the costs of its customer and thus increases the value it receives in the business relationship. However, only if the wholesaler can obtain a price that is sufficient to offset the cost of these services will this strategy be profitable. If this can be done we believe the strategy is a good one especially considering it is a strategy which is hard to copy quickly. A multiplex retailer could easily duplicate a pricing strategy but service related strategies are more hidden and thus harder to identify and replicate. In this regard distributors (both retailers and wholesalers) may find that they cannot meet global price competition, however, if they can provide value-added services more cost effectively and/or at a higher quality then they may have the means to remain competitive. The sophisticated business purchaser is looking at more than price but total value obtained in the business relationship.
Conclusion
Intertype competition has increased in retailing as retailers from different lines of trade have pursued scrambled merchandising and thus have found themselves competing with each other. More recently a new type of competition has evolved and is beginning to diffuse globally. This is where a new type middleman, which we call a multiplex retailer, caters to both the traditional customer of the wholesaler as well as to the household market the traditional retailer has historically pursued. As these multiplex retailers develop alternate distribution channels, retailers and wholesalers will find themselves increasingly competing with each other. Not only will retailers sell more to businesses but wholesalers will begin to pursue the household market. Given this development it is important that distribution educators and executives begin to better understand what will determine patronage preferences at these new multiplex business formats. This research has made a first attempt at this goal and it is hoped will stimulate additional research and discussion on this topic around the world. A sample of the research questions that might be addressed include:
How do business purchasers view the total value of purchasing model when they source products via mail order or the Internet?
How will the membership warehouse clubs, hardware home centers, and office supply superstores make inroads into rapidly developing countries such as China?
If power retailers can change how business sources products what does this mean for channel leadership? Can powerful multiplex retailers offer more value by offering private brands? In this regard will global private brands develop? If so how do conventional manufacturer's brands compete and survive?
Can manufacturers, using the total value of purchasing model, develop strategies to sell direct to business users and households? In short, which distribution intermediaries can be eliminated from the distribution system while at the same time increasing total value to the end user?
All of these issues are more than local in nature; they apply to more than the USA and Europe where mega-stores and power retailers have a good foothold. The challenge is global because new-wave distribution formats will continue to diffuse rapidly around the world as electronic commerce and world capital markets pursue innovations in distribution that offer competitive advantage.
References
1. Chanil, D. (1994), "Wholesale clubs: a new era?", Discount Merchandiser, Vol. 34 No. 11, pp. 38-51.
2. Darden, W. and Lusch, R. (1983), Patronage Behavior and Retail Management, North-Holland, New York, NY.
3. Doyle, P., Woodside, A.G. and Michell, P. (1979), "Organizations buying in new task and rebuy situations", Industrial Marketing Management, Vol. 8 No. 1, pp. 7-11.
4. Ellram, L.M. (1994), "A taxonomy of total cost of ownership models", Journal of Business Logistics, Vol. 15 No. 1, pp. 171-91.
5. Ellram, L.M. (1995), "Total cost of ownership: an analysis for purchasing", International Journal of Physical Distribution and Logistics Management, Vol. 25 No. 8, pp. 5-23.
6. Ellram, L.M. and Siferd, S.P. (1993), "Purchasing: the cornerstone of the total cost of ownership concept", Journal of Business Logistics, Vol. 14 No. 1, pp. 163-84.
7. Ferguson, T.W. (1997), "A revolution that has a long way to go", Forbes, Vol. 160 No. 3, pp. 106-12.
8. Giunipero, L.C. and Brewer, D.J. (1993), "Performance based evaluation systems under total quality management", International Journal of Purchasing and Materials Management, Vol. 29 No. 1, pp. 35-41.
9. Gregory, R.E. (1986), "Source selection: a matrix approach", Journal of Purchasing and Materials Management, Vol. 22 No. 2, pp. 24-9.
10. Kauffman, R.G. (1994), "Influences on industrial buyer's choice of products: effects of product application, product type, and buying environment", International Journal of Purchasing and Materials Management, Vol. 30 No. 2, pp. 29-38.
11. Lehman, D.L. and O'Shaughnessy, J. (1974), "Difference in attribute importance for different industrial products", Journal of Marketing, Vol. 38 No. 1, pp. 36-42.
12. Lehman, D.L. and O'Shaughnessy, J. (1982), "Decision criteria used in buying different categories of products", Journal of Purchasing and Materials Management, Vol. 18 No. 1, pp. 9-14.
13. Lusch, R.F. and Zizzo, D. (1995), Competing for Customers: How Wholesaler-Distributors Can Meet the Power Retailer Challenge, Distribution Research & Education Foundation, Washington, DC.
14. Maltz, A.B. and Ellram, L.M. (1997), "Total cost of relationship: an analytical framework for the logistics outsourcing decision", Journal of Business Logistics, Vol. 18 No. 1, pp. 45-66.
15. Nunnally, J.C. (1967), Psychometric Theory, McGraw-Hill, New York, NY.
16. Peter, P.J., Churchill Jr, G.A. and Brown, T.J. (1993), "Caution in the use of difference scores in consumer research", Journal of Consumer Research, Vol. 19 No. 1, pp. 655-62.
17. Puto, C.P., Patton, W.E. and King, R.K. (1985), "Risk handling strategies in industrial vendor selection decisions", Journal of Marketing, Vol. 49 No. 1, pp 89-98.
18. Smytka, D.L. and Clemens, M.W. (1993), "Total cost supplier selection model: a case study", International Journal of Purchasing and Materials Management, Vol. 29 No. 1, pp. 12-49.
19. Stern, L.W., El-Ansary, A.I. and Coughlan, A.T. (1996), Marketing Channels, Prentice Hall, Upper Saddle River, NJ.
20. Timmerman, E. (1986), "An approach to vendor performance evaluation", Journal of Purchasing and Materials Management, Vol. 22 No. 4, pp. 2-8.
21. Thompson, K.N. (1990), "Vendor profile analysis", Journal of Purchasing and Materials Management, Vol. 26 No. 1, pp. 11-18.
22. Thornton, E. (1994), "Revolution in Japanese retailing", Fortune, Vol. 129 No. 3, pp. 52-6.
23. Wilson, E.J. (1994), "The relative importance of supplier selection criteria: a review and update", International Journal of Purchasing and Materials Management, Vol. 30 No. 3, pp. 34-41.
Further reading
24. Anderson, E., Chu, W. and Weitz, B. (1987), "Industrial purchasing: an empirical exploration of buyclass framework", Journal of Marketing, Vol. 51 No. 3, pp. 71-86.
25. Bellizzi, J.A. and McVey, P. (1983), "How valid is the buygrid model?", Industrial Marketing Management, Vol. 12 No. 1, pp. 57-62.
26. Bonoma, T.V., Zaltman, G. and Johnston, W.J. (1978), Industrial Buying Behavior, Marketing Science Institute, Cambridge, MA.
27. Chao, C., Scheuing, E.E. and Ruch, W.A. (1993), "Purchasing performance evaluation: an investigation of different perspectives", International Journal of Purchasing and Materials Management, Vol. 29 No. 3, pp. 33-9.
28. Choffray, J. and Lilien, G. (1978), "Assessing response to industrial marketing strategy", Journal of Marketing, Vol. 42 No. 1, pp. 20-31.
29. Ferguson, W. (1979), "An evaluation of the BUYGRID framework", Industrial Marketing Management, Vol. 8 No. 1, pp. 40-4.
30. Jackson, D.W., Keith, J.E. and Burdick, R.K. (1984), "Purchasing agents' perceptions of industrial buying center influence", Journal of Marketing, Vol. 48 No. 4, pp. 75-83.
31. Johnston, W.J. (1994), "Organizational buyer behavior - 25 years of knowledge and research", Journal of Business and Industrial Marketing, Vol. 9 No. 3, pp. 4-5.
32. Kohli, A. (1989), "Determinants of influence in organizational buying: a contingency approach", Journal of Marketing, Vol. 53 No. 3, pp. 50-65.
33. Mayer, W.U. (1983), "Situational variables and industrial buying", Journal of Purchasing and Materials Management, Vol. 19 No. 3, pp. 21-5.
34. McQuiston, D.H. (1989), "Novelty, complexity, and importance as causal determinants of industrial buyer behavior", Journal of Marketing, Vol. 53 No. 1, pp. 66-79.
35. Moriarty, R.T. (1983), Industrial Buying Behavior, Lexington Books, Lexington, MA.
36. Moriarty, R.T. and Galper, M. (1978), Organizational Buying Behavior: A-State-of-the-Art Review and Conceptualization, Marketing Science Institute, Cambridge, MA.
37. Nauman, E., Lincoln, D.J. and McWilliams, R.D. (1984), "The purchase of components: functional areas of influence", Industrial Marketing Management, Vol. 13 No. 1, pp. 113-22.
38. Pingy, J.R. (1974), "The engineer and purchasing agent compared", Journal of Purchasing, Vol. 10 No. 4, pp. 33-45.
39. Qualls, W.J. and Puto, C.P. (1989), "Organizational climate and decision framing: an integrated approach to analyzing industrial buying decisions", Journal of Marketing Research, Vol. 26 No. 2, pp. 179-92.
40. Robinson, P.J., Faris, C.W. and Wind, Y. (1967), Industrial Buying and Creative Marketing, Allyn & Bacon, Boston, MA.
41. Ronchetto, J.R., Hutt, M.D. and Reigen, P.H. (1989), "Embedded influence patterns in organizational buying systems", Journal of Marketing, Vol. 53 No. 4, pp. 51-62.
42. Sheth, J.N. (1973), "A model of industrial buyer behavior", Journal of Marketing, Vol. 37 No. 4, pp. 50-6.
43. Sweeney, T.W., Mathews, H.L. and Wilson, D.T. (1973), "An analysis of industrial buyer's risk reduction behaviors: some personality correlates", in American Marketing Association Proceedings, American Marketing Association, Chicago, IL, pp. 217-21.
44. Webster, F.E. (1991), Industrial Marketing Strategy, John Wiley, New York, NY.
45. Webster, F.E. and Wind, Y. (1972), "A general model for understanding organizational buying behavior", Journal of Marketing, Vol. 36 No. 2, pp. 12-19.
| [Illustration] |
| Caption: Table I; Normative models of supplier selection; Table II; Total cost/value service component comparison; Figure 1; Total value of purchasing model; Table III; Categorization of elements of total value model; Table IV; Reliability of difference scales[sup]a; Table V; Regression results[sup]a |
My Research