You are on page 1of 7

A consumer model for channel switching behavior

James Reardon and Denny E. McCorkle

Introduction
According to the Direct Marketing Association, consumer direct order marketing sales are growing faster than overall US consumer sales. From 1995 to 2000, these sales increased by 9.1 percent annually (compared to 5.5 percent overall), with an 8.3 percent annual growth expected through 2005 (compared to 4.1 percent overall). With direct mail (including catalogs) leading the media for direct order sales (34.8 percent), interactive sales (including Internet) are forecast at $24.2 billion in 2000 and are projected to increase 41.3 percent per year to reach $36.4 billion by 2005 (Direct Marketing Association, 2001). Many have attributed this continuing growth in direct order marketing to its relative time savings and convenience compared to traditional retailing channels (Darian, 1987; Eastlick and Feinberg, 1994; Gehrt et al., 1996; Lavin, 1993). More specifically, recent industry studies have focused on the use of alternative distribution channels and found an increasing use of the Internet as a communication channel for information and comparison shopping, and as a distribution channel for online purchasing (NPD Group, Inc., 2001; NRF 2000). Although there is a large body of research investigating consumers' store patronage behavior, there has been a lack of theoretical development to adequately explain consumers' channel-switching behavior between traditional store and direct order marketing channels (Darian 1987; Gehrt and Carter 1992). This study attempts to build a theoretical model that explains consumers' choice of distribution channels by extending Becker's (1965) time allocation model. Under Becker's model, a household is assumed to allocate optimal amounts of time and money income on a series of household activities to achieve the maximum level of utility. Often shopping is considered a transaction cost instead of a household activity. However, the shopping experience has the hallmarks of household production (i.e. goods and services produced by household members) in that household labor (i.e. time by member of the family) can, to a certain degree, be substituted for market services (e.g. personal shoppers, delivery). Based on the extended Becker's model, consumers are assumed to choose a

The authors James Reardon is Wells Fargo Professor of Marketing at the University of Northern Colorado, Greeley, Colorado, USA. Denny E. McCorkle is Professor of Marketing, Southwest Missouri State University, Springfield, Missouri, USA. Keywords Consumer behaviour, Marketing, Distribution channel Abstract With the phenomenal growth of direct order marketing with the Internet and catalogs as alternative channels, customers increasingly face more choices of where to purchase goods and services. This paper develops a formal consumer model to explain channel switching behavior. Becker's theory of time allocation is expanded to consumer decision making between distribution channels. The final model suggests that consumers face a tradeoff when deciding where to buy goods and services. From this tradeoff an indifference curve is developed where the consumer chooses between alternative distribution channels on the basis of the relative opportunity costs of time, costs of goods, pleasure derived from shopping, perceived value of goods, and relative risk of each channel. Strategies for direct and multi-channel marketers are developed using this model. Electronic access The research register for this journal is available at http://www.emeraldinsight.com/researchregisters The current issue and full text archive of this journal is available at http://www.emeraldinsight.com/0959-0552.htm
International Journal of Retail & Distribution Management Volume 30 . Number 4 . 2002 . pp. 179185 # MCB UP Limited . ISSN 0959-0552 DOI 10.1108/09590550210423654

179

A consumer model for channel switching behavior

James Reardon and Denny E. McCorkle

International Journal of Retail & Distribution Management Volume 30 . Number 4 . 2002 . 179185

distribution channel that gives them the maximum utility for minimum input of household resources. Becker's model is particularly suited to explain channel-switching behavior. In essence, a consumer's choice between commodities is largely one of transaction cost to the consumer. This transaction cost involves all the fundamental elements of Becker's model of time allocation: household time, resource availability, the implication of psychic costs/benefits, and capital constraints.

Modeling household time allocation


Becker's (1965) theory of time allocation is referred to as an integration of the theory of the consumer with that of the firm (Deaton and Muellbauer, 1980). Mathematically, a household obtains utility from the underlying commodities, Z, which are produced using market goods and time as inputs (Michael and Becker 1973). Thus: U = u (Z). (1) The technology, by which market goods X are used in combination with time and capital K, is represented as a household production function: (2) Z = z (X, Lo; K) where Lo is a household's time spent in producing Z. The utility function (equation (1)), is maximized subject to the production function constraints (equation (2)), and a constraint on the household's available time and income: (3) T = Lw + Lo I = PX (4) where T is the total time available, and Lw is the household's time spent in the labor market. I is income, and P is a vector of the price of market goods. The time and money income constraints can be collapsed into a single resource constraint on the household's ``full income'', or S (Becker, 1965). Thus: S = Wo T + V = Wo (Lw + Lo) + (5) V = I + W o Lo where Wo is a vector of wage rate of the household member and V is unearned income. The utility function (equation (1)), is maximized subject to the constraints of the production functions (equation (2)), and full income (equation (5)). In solving this optimization problem, Deaton and Muellbauer (1980) proposed a two-stage approach. At the lower stage, the cost of producing the vector Z is minimized, just as for the firm. The objective is to minimize the cost at a given technology K and Z. The solution is a function: (6) C = c(P, Wo,Z;K). The upper stage of the household optimization problem is then to choose Z to maximize utility: U = u(Z) such that C = c(P, Wo,Z;K). (7)

Literature review
It has long been recognized that service retailers should examine customers as vital participants in the transaction process (Mills, 1986). Hollander (1964) recognized that marketers could shift work to others upward or downward in the distribution channel. Despite the early recognition that retailers may shift work to consumers, there is still relatively little research on the area of work shifting between consumers and channel members. The decision to purchase can be viewed as a resource allocation decision. According to Nobel Laureate, Gary S. Becker, ``the household is truly a `small factory': it combines capital goods, raw materials and labor to clean, feed, procreate and otherwise produce useful commodities'' (Becker, 1965, p. 496). As such, the decision about which channel to use to purchase goods and services is essentially a household production decision, because household members allocate time and resources in the purchase decision. While it appears logical that the choice of shopping channel choice fits within this model, there is no research to date that adequately makes use of Becker's theory to model channel choice. Although the importance of household production was recognized by early marketing scholars (Alderson, 1958; Bartels, 1944), there is a lack of scholarly research pertaining to the subject of household production and subsequent consumption. This is particularly true with regard to the expenditure of household time in the acquisition activities.

180

A consumer model for channel switching behavior

James Reardon and Denny E. McCorkle

International Journal of Retail & Distribution Management Volume 30 . Number 4 . 2002 . 179185

Then, the demand function for household produced commodities, Z, can be derived as follows: Z = z(S, P, Wo;K). (8) Several consumer researchers have built parallel to Becker's (1965) theory by examining the time allocation of consumers in the purchase process (e.g. Etgar, 1978; Nickols and Fox, 1983; Strober and Weinberg, 1977, 1980; Feldman and Hornik, 1981; Lusch et al., 1992). While using some of the same antecedents, these models are largely tangential to the current research.

Development of an extended model


Consumers often substitute buying a good through one distribution channel with another channel because of the relative household time and effort involved (Darian, 1987; Eastlick and Feinberg, 1994). According to Kelley et al. (1990), consumers actually participate in the transaction process at the retail level by acting as ``partial employees''. That is, consumers trade-off time and energy to achieve better quality or lower prices at the retail level (e.g. self-service at Wal-Mart). Consumers are willing to trade their time and effort in exchange for a value differential. Other retailers make other trades with consumer time. For example, entertainment may be an integral part of the shopping experience. Many malls are increasing expenditures to encourage customers to drive further and stay longer within the mall. In a similar way, many of today's direct order catalogs are designed for leisure reading and viewing, with more than just direct ordering in mind (Gehrt and Carter, 1992; Greene, 1984; Shim and Mahoney, 1992). Also, many Internet Web sites contain entertaining and helpful content to encourage repeat visits and create a desire of consumers to trade whatever it is they get from other channels to the Internet. Although not separately modeled, Becker (1965) suggests that consumers obtain process value from conducting household activities in the form of psychic income. Consumers obtain psychic income, or pleasure, from the shopping experience. Thus, consumers obtain utility from shopping both in terms of utility from their purchases and the pleasure of shopping. In order to

obtain these utilities, consumers input shopping and transaction time, money, and the use of capital, such as automobiles. The goal of the rational consumer is to maximize the utility from shopping subject to cost and full income constraints. The cost function of shopping is the opportunity cost of time, the price of the product, the quantity of products, and subject to some capital usage (e.g. Deaton and Muellbauer, 1980). The choice for a consumer to choose one distribution channel over another can be viewed as an optimization problem. Assuming a two-channel system, one traditional retailer (I) and one direct order marketer (j), it can be shown that consumers will switch between channels when the utilities derived from using one channel relative to the costs involved outweigh the same for an alternative channel, subject to the full income and capital constraints. Following Becker's logic, the utilities derived from shopping at a traditional retailer versus a direct marketer include the relative pleasure from the shopping experiences, relative prices and quantity purchased, and differentials in the time involved in the shopping process. These utilities are maximized at a given income, wage rate, and available capital, assuming the wage rate as an opportunity cost of time (Becker, 1965). If capital is freely available or a low cost viable substitute is available, the capital constraint will not be considered in a marginal decision. However, if there are capital constraints on the shopping process (e.g. lack of transportation to the store, lack of access to a computer to shop on the Internet) this is likely to have a large differential effect. For simplicity, we assume that there is no marginal effect of capital on the choice of channels. Likewise, since full income is relatively constant in both circumstances, we can assume no marginal effect of the full income variable except to the extent that price and time usage have in changing future behavior. For example, if a consumer buys from a direct order marketer and saves two hours of time, this time can then be allocated to leisure or work. The value of this time is taken into consideration with the time costs associated with the model. Since consumer decisions are made at the margin, we can model channel choice by optimizing the marginal benefit subject to costs. The relative marginal costs, RMC, of

181

A consumer model for channel switching behavior

James Reardon and Denny E. McCorkle

International Journal of Retail & Distribution Management Volume 30 . Number 4 . 2002 . 179185

using one channel to the exclusion of the other is: RMC = Q(Pi-Pj) + wage (Ti-Tj) (9) given that the customer is buying Q number of products at P price levels and inputting T amount of time in the shopping process. The first part of the equation, [Q(Pi-Pj)], represents the direct money cost to the consumer, while the second part, [wage (Ti-Tj)], represents the indirect opportunity cost of the consumer's time. Likewise, if we assume that the same commodities are purchased through either channel, the relative marginal benefit RMB to the consumer can be shown as: RMB = PIi Pij. (10) With the assumption of exogenous demand, the only difference in benefit between the distribution channels becomes the psychic income or pleasure component PI from shopping. The indifference curve between the two shopping channels can be constructed at a point where the relative difference in the net gain between the two channels is zero. Thus, the consumers will be indifferent between channel choice at: (PIi PIj) [Q (Pi Pj) + wage (Ti Tj)] = 0. (11) Although this model provides a good start to examining the process of distribution channel choice, it is restrictive since the purchased products are assumed to be singular commodities. Furthermore, although the shape of the utility function (strictly quasiconcave) assumes a risk adverse consumer, the previous model does not allow differential risks between channels to be considered. The model becomes more accurate and practical by relaxing these assumptions. If we first relax the assumption that the products are commodities, allowing for product differences such as brands and quality differences, we extend the benefit side to include any differential perceptions of value V and integrate this into the indifference curve as follows: [Q (Vi Vj) + (PI I PI j)] [Q (Pi Pj) + wage (Ti Tj)] = 0. (12) Research has indicated that the risk of a purchase can be transferred to the retailer that sells it (Dash et al., 1976; Hisrich et al., 1972) and, more specifically, that customers may perceive differing levels of risk in shopping

through different channels (Akaah and Korgaonkar, 1988; Darian, 1987; Schiffman et al., 1976; Simpson and Lakner, 1993; Van den Poel and Leunis, 1996). This risk can be modeled by including a risk of loss of the price of a product through the perceived probability of product failure, R, or non-delivery: [Q (Vi Vj) + (PI I PI j)] [Q (Pi Pj) + wage (Ti Tj) + (Ri(Pi) Rj(Pj))] = 0. (13) The assumption that only a single good is bought on a shopping trip is somewhat limiting. Instead, a comparative bundle of goods, 1. . .k, can be integrated to make the model more realistic. Since we still assume a single shopping trip for comparison, the psychic income and opportunity cost components do not change. These two components denote the value or cost of the transaction process as a whole, including access time (e.g. travel time) and shopping time. Thus, the resultant indifference curve of integrating a basket purchase is: [{Qk (Vik Vjk)} + (PIi PIj)] [{Qk (Pik Pjk)} + wage (Ti Tj) + {(Rik(Pik) Rjk(Pjk)})] = 0 (14) where j represents a traditional retailer and i a direct order marketer. The first component in square brackets is the relative marginal benefits and the second the relative marginal costs to the consumer. The relative marginal benefits include differences in perceived value {Qk (Vik Vjk)} and psychic income (PIi PIj). The relative marginal costs include the cost of the goods {Qk (Pik Pjk)}, the opportunity cost of the consumer's time wage (Ti Tj), and the relative risk of loss {Rik(Pik) Rjk(Pjk)} .

Implications and discussion


There are two major sets of implications from this model: theoretical and managerial. The theoretical implications arise from the extension of Becker's (1965) model into the domain of partial household production. While the theory of time allocation has been examined in some depth and shown to be robust across situations, it needs further examination in the current context. Also, further extensions are not limited to the ones given above.

182

A consumer model for channel switching behavior

James Reardon and Denny E. McCorkle

International Journal of Retail & Distribution Management Volume 30 . Number 4 . 2002 . 179185

The objective of this paper is to provide a basic model that can be adapted to shopping channel choice. Indeed, in the context of the current e-tailing rage and the predicted doom of traditional retailing, the introduction of a general model to examine the channel choice of consumers seems timely. In fact, the more recent dot-bomb phenomena of e-tailers going out of business, especially after all the media hype, would seem a fertile field in which to test this model. To date, few if any models have been proposed that allow for the empirical examination of channel-switching behavior from the consumer viewpoint. The managerial implications for this model are numerous, and include those discussed in the following subsections. A consumer's channel-switching behavior is influenced by tradeoffs One such tradeoff is between time and money. Direct order marketing channels offer the advantage of time savings (e.g. traveling to and from the store and shopping time), but a disadvantage concerning time when a purchase needs returning. While direct order shopping through the Internet can often result in better prices, the total cost including shipping and handling often erase such money savings. This money savings will diminish further when the current moratorium on sales tax is lifted. Nonetheless, in its current state, an experienced Web consumer can also save money if they take the time to use comparison shopping bots (e.g. MySimon.com) and to benefit from the frequent Internet marketing practice of offering substantial discounts to acquire new Web customers. Though, as Web marketers are pressed for showing profits from their investors, their strategies are more likely to follow their catalog channel competitors and compete more on the basis of product availability and convenience rather than strictly on price. Another tradeoff is between time and psychic income. Regardless of the time or money saved by shopping from the Web or from catalogs, some consumers will continue to desire a shopping experience that extends beyond the utility of it. These consumers will view shopping in some product categories as a pleasurable, perhaps even a social, experience where saving time is not their primary objective. For this, depending on the product category and the individual, traditional

retailers may hold the advantage and place the direct order marketers in need of strategies for improved competition in this area. However, for other more technology-oriented consumers, pleasure derived from shopping on the Internet may be greater than from shopping through other channels. Some Internet marketers have successfully used pleasure or psychic-enhancing strategies such as community building, content publishing, and special events. Community building is where customers and Web site visitors are encouraged to interact and socialize with each other through live chat and postings on discussion boards about their special interests and shopping experiences with the community sponsor or marketer (e.g. Expedia.com, Travelocity.com). Besides this interactive content, some Web marketers also provide additional content of special interest to their target market. This content publishing often provides a wealth of information and expertise to assist the Web shopper in making a more informed and confident decision (e.g. Amazon.com, LLBean.com). Another means of providing psychic benefits to Web users is through special events. For example, the music Web site CDNow.com regularly promotes live streaming concerts and interactive chats with their popular music artists. While certainly not physical, savvy Web marketers are finding virtual ways to add psychic benefits to the Web shopping experience, many to their benefit over other direct order marketing channel competitors such as catalogs. Though fortunately for catalogs, many have added a Web site as a supplemental channel and means of improving their interaction with customers. A consumer's channel-switching behavior is influenced by perceived risk While the risks of shopping can be associated with traditional store retailers, these risks are increased through channels operating from a distance. Besides perceived financial risk, other risks such as performance, social, frontend time-loss, back-end time-loss, and source risk, have been proposed for situations of direct order shopping (McCorkle, 1990). Some, if not all, of these risks may need addressing by direct order marketers in order to compete with their physical store counterparts. For example, financial risks can be reduced by promoting money-back

183

A consumer model for channel switching behavior

James Reardon and Denny E. McCorkle

International Journal of Retail & Distribution Management Volume 30 . Number 4 . 2002 . 179185

guarantees and their added value of convenient and helpful live customer service by phone or online chat. Performance risk and back-end time-loss risk is reduced when returns can be made at physical stores or through physical partners. Front-end timeloss risk is reduced with a clear presentation of product inventory availability and shipping/ delivery times. Source risk is reduced with seals of approval by outside organizations such as the Better Business Bureau, BizRate, and TRUSTe; by developing and supporting privacy policies and permission-based marketing practices; and with continued assurances and technology for secure transactions and safeguarding of personal data such as credit card and order information. Many customers perceive higher levels of security and privacy violations from direct order marketers (Easley et al., 1992; Garman, 1996; Laczniak et al., 1995; Phelps and Bunker, 2001; Sheehan, 1999). The problem of channel-switching behavior can be addressed through use of multi-channel marketing strategies Because a consumer is likely to vary their shopping behavior by channel, perhaps the strongest implication of our extension of Becker's (1965) model is that the greatest success will flow to the retailer with the greatest channel choice available. The retailer with multiple channel options (e.g. physical store, catalog, Web, and/or wireless), will be the retailer that can best serve the consumer's need for communication of information and distribution of goods and services. Channelswitching behavior becomes less of a problem for retailers providing multiple channels for their customers and Becker's model provides a manner to understand and predict this behavior.

References
Akaah, I.P. and Korgaonkar, P.K. (1988), ``The influence of product, manufacturer, and distributor characteristics on consumer interest in direct marketing offerings'', Journal of Direct Marketing, Vol. 3, Summer, pp. 27-33. Alderson, W. (1958), ``The analytical framework for marketing'', in Duncan, D.J. (Ed.), Proceedings of the Conference of Marketing Teachers from Far Western States, University of California, Berkeley, CA, pp. 15-28. Bartels, R.D. (1944), ``Marketing principles'', Journal of Marketing, Vol. 9, October, pp. 151-7.

Becker, G.S. (1965), ``A theory of the allocation of time'', The Economic Journal, Vol. 75, September, pp. 493-517. Darian, J.C. (1987), ``In-home shopping: are there consumer segments?'', Journal of Retailing, Vol. 63 No. 2, pp. 163-86. Dash, J., Schiffman, L.F. and Berenson, C. (1976), ``Risk and personality-related dimensions of store choice'', Journal of Marketing, Vol. 40, January, pp. 32-9. Deaton, A. and Muellbauer, J. (1980), Economics and Consumer Behavior, Cambridge University Press. Direct Marketing Association (2001), Economic Impact: US Direct Marketing Today, Direct Marketing Association, New York, NY, http://www.the-dma. org/library/publications/libres-ecoimpact2.shtml Easley, R.W., Roberts, J.A., Dunn, M.G. and Madden, C.S. (1992), ``Diagnosing consumer information problems: an investigation of deception in the mailorder video camcorder market'', Journal of Public Policy and Marketing, Vol. 11 No. 2, pp. 37-44. Eastlick, M. and Feinberg, R. (1994), ``Gender differences in mail-catalog patronage motives'', Journal of Direct Marketing, Vol. 8 No. 2, pp. 37-44. Etgar, M. (1978), ``The household as a production unit'', in Sheth, J. (Ed.), Research in Marketing, Vol. 1, JAI Press, Greenwich, CT, pp. 79-98. Feldman, L.P. and Hornik, J. (1981), ``The use of time: an integrated conceptual model'', Journal of Consumer Research, Vol. 7, March, pp. 407-19. Garman, E.T. (1996), Ripoffs and Frauds: How to Avoid and How to Get Away, Dame Publications, Houston, TX. Gehrt, K.C. and Carter, K. (1992), ``An exploratory assessment of catalog shopping orientations: the existence of convenience and recreational segments'', Journal of Direct Marketing, Vol. 6, Winter, pp. 29-39. Gehrt, K.C., Yale, L.J. and Lawson, D.A. (1996), ``The convenience of catalog shopping: is there more to it than time?'', Journal of Direct Marketing, Vol. 10, Autumn, pp. 19-28. Greene, R. (1984), ``A boutique in your living room'', Forbes, 7 May, pp. 86-94. Hisrich, R.D., Dornapf, R.J. and Kernan, J.B. (1972), ``Perceived risk in store selection'', Journal of Marketing Research, Vol. 9, November, pp. 435-9. Hollander, S.C. (1964), ``Who does the work of retailing'', Journal of Marketing, Vol. 28, July, pp. 18-22. Kelley, S.W., Donnelly, J.H. Jr and Skinner, S.J. (1990), ``Customer participation in service production and delivery'', Journal of Retailing, Vol. 66 No. 3, pp. 315-35. Laczniak, R.N., Muehling, D.D. and Carlson, L. (1995), ``Mothers' attitude toward 900-number advertising directed at children'', Journal of Public Policy and Marketing, Vol. 14 No. 1, pp. 108-16. Lavin, M. (1993), ``Wives' employment, time pressure, and mail/phone order shopping: an exploratory study'', Journal of Direct Marketing, Vol. 7 No. 1, pp. 42-9. Lusch, R.F., Brown, S.W. and Brunswick, G.J. (1992), ``A general framework for explaining internal vs external exchange'', Journal of the Academy of Marketing Science, Vol. 20 No. 2, pp. 119-34. McCorkle, D.E. (1990), ``The role of perceived risk in mail order catalog shopping'', Journal of Direct Marketing, Vol. 4, Autumn, pp. 26-35.

184

A consumer model for channel switching behavior

James Reardon and Denny E. McCorkle

International Journal of Retail & Distribution Management Volume 30 . Number 4 . 2002 . 179185

Michael, R.T. and Becker, G.S. (1973), ``On the new theory of consumer behavior'', Swedish Journal of Economics, Vol. 75, pp. 378-96. Mills, P.K. (1986), Managing Service Industries, Ballinger Publishing Co., Cambridge, MA. Nickols, S.Y. and Fox, K.D. (1983), ``Buying time and saving time: strategies for managing household production'', Journal of Consumer Research, Vol. 10, September, pp. 197-208. NPD Group, Inc. (2001), ``NPD e-Visory report shows offline sales benefit from online browsing'', NPD Online Research, 17 July, http://www.npd.com/ corp/content/news/releases/press_010717.htm NRF (2000), ``Joint NRF/J.C. Williams Group/Bizrate.com Channel Surfing Study validates cross-channel influence'', National Retail Federation, 25 September, http://www.nrf.com/content/ default.asp?folder=press&file=sept2500.htm Phelps, J.E. and Bunker, M.D. (2001), ``Direct marketers' use of public records: current legal environment and outlook for the future'', Journal of Interactive Marketing, Vol. 15, Winter, pp. 33-48. Schiffman, L.G., Schus, S. and Winer, L. (1976), ``Risk perception as a determinant of in-home

consumption'', Journal of the Academy of Marketing Science, Vol. 4, Fall, pp. 753-63. Sheehan, K.B. (1999), ``An investigation of gender differences in on-line privacy concerns and resultant behaviors'', Journal of Interactive Marketing, Vol. 13, Autumn, pp. 24-38. Shim, S. and Mahoney, M.Y. (1992), ``The elderly mailorder catalog user of fashion products: a profile of the heavy purchaser'', Journal of Direct Marketing, Vol. 6 No. 1, pp. 49-58. Simpson, L. and Lakner, H.B. (1993), ``Perceived risk and mail order shopping for apparel'', Journal of Consumer Studies and Home Economics, Vol. 17, pp. 377-89. Strober, M.H. and Weinberg, C.B. (1977), ``Working wives and major family expenditure'', Journal of Consumer Research, Vol. 4 No. 1, pp. 141-7. Strober, M.H. and Weinberg, C.B. (1980), ``Strategies used by working and nonworking wives to reduce time pressures'', Journal of Consumer Research, Vol. 6 No. 4, pp. 338-48. Van den Poel, D. and Leunis, J. (1996), ``Perceived risk and risk reduction strategies in mail-order versus retail store buying'', The International Review of Retail, Distribution and Consumer Research, Vol. 6, October, pp. 351-71.

185