Ann Oper Res DOI 10.1007/s10479-014-1781-5

Green, greener or brown: choosing the right color of the product Arda Yenipazarli · Asoo J. Vakharia

© Springer Science+Business Media New York 2015

Abstract There does not exist a one-size-fits-all green product strategy in green market, and formulating a strategy that aligns a company’s economic and environmental goals is no small feat. In this paper we provide insights into a firm’s green product strategy choice with a focus on two alternative strategies: Greening-Up and Greening-Out. In doing so, we incorporate two important characteristics of the customer market where there are substantial numbers of customers who are potentially receptive to a green appeal: (1) Customer market is divided into three distinct and mutually exclusive segments based on large-scale surveys and interviews conducted to measure both customers’ willingness-to-pay for products with environmental attribute(s) and their propensity to buy these products; and (2) We factor in the findings of the most recent market behavior studies that even the customers that demonstrate the least environmental responsibility of all the segments can buy green products for non-environmental reasons. The contributions and findings of our work are as follows. First, we characterize the market- and product-related factors a firm should act on for a successful execution of green product strategies. Second, on the basis of customer choice data available, we assess the strategic fit of distinct pricing options under each strategy choice with the objective(s) of a firm. Third, we explore the extent to which a green product strategy creates higher environmental benefits while providing economic payoffs to a firm. In contrast to prevailing view, we show that greening up an existing brown product is not necessarily better at reducing the environmental impact of a firm than designing a new green product. We observe that responsiveness of the least environmentally conscious customers to environmental attributes added into a brown product sets one of the two major constraints on the environmental performance of Greening-Up strategy. We also observe that there does

A. Yenipazarli Department of Logistics and Supply Chain Management, Georgia Southern University, Statesboro, GA 30460, USA e-mail: [email protected] A. J. Vakharia (B) Department of Information Systems and Operations Management, University of Florida, Gainesville, FL 32611, USA e-mail: [email protected]

123

Ann Oper Res

not exist a strict dichotomy between having a better economic performance and achieving a higher environmental performance, and a firm can achieve both goals simultaneously, even if it means leaving out serving some of the customers targeted by the firm’s existing brown product. Keywords Environmental sustainability · Product positioning · Portfolio design · Market segmentation

1 Introduction Not too long ago, green shoppers in the marketplace were only a minority with inadequate buying power to draw attention to green products. At that time, green shopping was viewed as part of a fringe lifestyle and green products could not find their ways into the shopping lists. Green products did not even appear in firms’ To-Do lists because environment was not the focus of public attention, and most firms tended to follow the old phrase “If it ain’t broke, don’t fix it.” The widespread thinking was “this is a bad thing for our bottom line.” Products that claim to be “green” were mostly from tiny mom-and-pop firms and they had all gathered dust on the bottom shelves of the stores, if not forgotten at all in back rooms or basements, because they used to carry a negative baggage. They did not perform as well as their “brown” counterparts and were not a good value. That is all water under the bridge. Today, there are substantial numbers of consumers who are espousing environmental values and changing their shopping lists. The waves of consumer desires for green products have finally started to earn mainstream understanding to drive action. For instance, more than one-third of the U.S. consumers are purchasing natural cleaning/personal care products, organically produced foods, compact-fluorescent lamps, and rechargeable batteries. Thanks in part to advances in technology and government mandates, a growing selection of today’s green products are alluring, highly accessible, and designed to deliver convenience, lower costs and perform better than their brown counterparts. For example, energy efficient, front-loading washers clean better and are gentler on clothes compared to conventional top-loading washers. This helps to broaden the appeal of green products to a wider customer audience. Energy-efficient electronics and appliances, coldwater and ultra-concentrated detergents, and air- and water-filtration devices are now on the top of consumers’ shopping lists. The U.S. green marketplace was reported to be almost $300 billion in 2010, representing investments in organic foods, Energy Star appliances, hybrids and electric vehicles, green home furnishings and apparel, and renewable power.1 Interest in green shopping is not just here to stay; it is also growing rapidly. The Natural Marketing Institute (NMI) predicted the green marketplace to reach $420 billion by 2010, and $845 billion by 2015. Even the ongoing economic woes have not deterred green sentiments in purchasing behavior, nor consumers’ expectations that companies will develop high-quality green products. Information Resources, Inc. found that consumers are actually maintaining or increasing their spending on green products in spite of the recently down economy.2 For instance, Seventh Generation’s sales grew by 20 % in 2007 and 50 % in 2008, while Clorox Green Works became the best-selling cleaning brand during its market debut year 2008.3 This 1 http://blog.lohas.com/blog/jacquelyn-ottmans-blog, accessed on May 16, 2014. 2 http://www.iriworldwide.com/NewsEvents/PressReleases/ItemID/749/View/Details.aspx,

accessed on May 16, 2014. 3 http://www.supermarketbattles.com/images/articles-pdf/042009.pdf, accessed on December 6, 2010.

123

Ann Oper Res

has led to green products being increasingly available on the shelves, and increasing numbers of shoppers are now putting green products into their shopping carts. Apparently, this shift in green behavior has been gaining forefront attention of companies, since they have been rolling out more and more green products—to seize the inherent opportunities available in green market. Toyota’s fuel-efficient Prius, Procter & Gamble’s Duracell rechargeable batteries, energy-saving Tide Coldwater laundry detergents and PUR water filtration products, Philips’ Marathon CFL bulbs, and Whirlpool’s Duet front-loading washers and dryers are just some of the notable examples of this trend. It used to be that consumers simply expected the products they would buy to perform well, be affordably priced, and be sold under familiar brand names. Apparently, that seemingly short and simple checklist now includes a highly varied record of environmental considerations. This presents to companies looking to address the sustainability-related product concerns of customers the need to improve the environmental performance of their products and processes, and restructure their agenda on the basis of environmental considerations. Klenier (1991) brings into focus three important questions that any company’s green agenda should include: (1) What type of a green product should a company bring to the market, and what materials should the firm include in it? (2) How should a company align his environmental actions with environmental attitudes and behaviors of consumers? and (3) How can a company address environmental issues in product design? Hopkins (2010) points out that a company recognized for his brown offerings can mould the perceptions of environmentally sensitive customers about his product(s) either by adding green features into the existing product or by coming up with a radically different design that is unabashedly better for the environment. These observations motivate the focus of this paper. In this paper we evaluate alternative green product strategies a firm can choose and implement for integrating environmental benefits into his product design. The specific research questions we address in this paper are as follows. First, in today’s green marketplace which is segmented by customers’ varying degrees of environmental concern and purchasing behavior, what are the key parameters that drive a firm to choose a specific strategy? Essentially, we are interested in identifying the key drivers which motivate the firm to introduce either a product which is completely environmentally friendly or a product which offers both traditional and environmentally friendly attributes. A second question relates to which strategy choice would be preferred from an environmental perspective and whether a profit-maximizing strategy choice might always be the one that results in greater environmental benefits. Finally, we also address the issue of market coverage in that whether a profit maximizing strategy choice does result in incomplete or complete market coverage. The remainder of this paper is organized as follows. Section 2 reviews relevant literature with a focus on product design and variety in vertically differentiated consumer markets. Section 3 describes our modeling framework, followed by a structural characterization of each strategy choice in Sect. 4. In Sect. 5, we discuss the economic and environmental consequences of each strategy choice, and drivers for and obstacles to implementing each strategy. Section 6 contains concluding remarks and directions for future research.

2 Literature review Our paper contributes to the growing literature on product-line design and sustainability. The management of product design and development has been extensively studied in marketing and operations management. Product-line design typically involves decisions of what product(s) to offer, how to position them in a heterogeneous consumer market and what pricing

123

Ann Oper Res

regime to apply. In this context, effective product design and associated pricing decisions are crucial to extract value from a heterogeneous population of customers. Product designs are defined by multiple attributes that can be grouped into two main categories: vertical and horizontal differentiation. Vertical differentiation captures the product performance or quality as a well-established driver of consumers’ willingness-to-pay, whereas horizontal differentiation represents the feature choices of product design. Our paper is centered around prior work on vertical differentiation and the reader is referred to Lancaster (1990) for a comprehensive review of earlier studies that deal with horizontally differentiated markets. Mussa and Rosen (1978), O’Keeffe (1980) and Katz (1984) are the first among related research papers that restrict market behavior into vertical differentiation via product quality and analyze price-quality decisions of a firm. In their seminal work, Mussa and Rosen (1978) derive a firm’s optimal price-quality schedule offered to a heterogeneous customer market with a continuous preference parameter along the quality dimension over a bounded range. Moorthy (1984) substantiates the benefits of serving multiple consumer segments through product design and different pricing schedules when consumers have discrete types. In this paper, the firm offers a menu of vertically differentiated products such that higher quality products are priced higher. As an alternative approach, Dobson and Kalish (1988) propose a model that determines how a firm should position/reposition and price a line of substitute products. Kim and Chhajed (2000) analyze the effects of commonality on customers’ valuation of products when it is used to design products in different classes. Common features increase similarity between products and influence customers’ choice of products, and they study the trade-off between cost-savings and revenue losses due to reduced product differentiation. Desai et al. (2001) use a model with two distinct customer segments and rule out such direct valuation changes and analyze alternative product design configurations with common components by introducing a model that investigates the trade-off between revenue losses resulting from reduced product differentiation and cost savings induced through design effort. By generalizing the models in Mussa and Rosen (1978) and Moorthy (1984), Kim and Chhajed (2002) study a firm’s product design problem with multiple quality-type attributes in a customer market partitioned into two distinctive segments. They derive a measure of multidimensional customer preference and show that a single product offering is never optimal for their two-segment setting. Choudhary et al. (2005) study the effect of personalized pricing on a firm’s quality choices. Based on complete knowledge of customers’ willingness-topay, the firm charges different prices to different consumers based on their willingness-topay. Heese and Swaminathan (2006) focus on a firm that designs a product line with two products, consisting of components that can be common for or unique to both products. The firm determines the component quality levels, and whether to use common or different components for the two products. For each component, the firm can reduce the cost of production by exerting a certain amount of effort. Instead of beginning with the premise that offering a product line is optimal and then just tackling the question of pricing, Anderson and Dana (2009) attempt to answer the question of whether price discrimination is always profitable and characterize the necessary conditions. Wu (2010) studies a firm that designs a single product with two quality dimensions, namely design quality and conformance quality, to serve consumers with diverse willingness-to-pay and maximize his profits. By integrating these two quality dimensions, the interrelationships between design quality, conformance quality, price and market demand are analyzed. Our paper also adds to the growing body of literature on green product design. In a study closely related to this paper, Chen (2001) proposes a utility model where a product is characterized by two quality-type attributes, namely traditional and environmental attributes, and analyzes a firm’s strategic decisions and policy issues regarding the design of a product

123

Ann Oper Res

line. By jointly considering the interactions among the preferences of customers, product strategies, and environmental standards imposed by governments, it is shown that green product development and stricter environmental standards might not necessarily benefit the environment. Recently, using a model similar in most respects to Chen (2001), Krishnan and Lecourbe (2010) focus on a firm that faces a trade-off between traditional and environmental attributes of his product(s) as well as the differential willingness-to-pay of customer market for these distinct quality dimensions. In this context, they study policies that promote sustainable innovation, under which it is possible for the firm to both maximize his profits and improve the aggregate environmental quality of his product line. More recently, Yenipazarli and Vakharia (2015) provide strategic insights into expanding a brown product line with a new green product. In their analysis, they explicitly incorporate a segmented consumer market where individual consumers may value the same product differently, the economies of scale and the learning effects associated with new green products, and capacity constraints for the current production system. In the context of green product design, there are also papers of peripheral relevance in remanufacturing literature, since remanufactured products are associated with lower environmental impact. Some of the key papers in this stream of literature are those of Ferrer (1996), Debo et al. (2005), Atasu et al. (2008), Ferrer and Swaminathan (2010), and Subramanian and Subramanyam (2012). Rather than reviewing this strand of literature, we refer the reader to Souza (2013) who offers an excellent review of extant contributions in this field. The key distinguishing features of our paper are as follows. First, customers are segmented into three distinct and mutually exclusive groups, characterized by their commitment to green, and involvement in green values and purchasing. The importance of this unique market structure is that it makes targeted marketing of consumers with different environmental attitudes possible. This segmentation model is drawn directly upon the recent Green Gauge reports prepared by the GfK Roper Consulting, which examines how U.S. adult consumers’ environmental knowledge affects their purchasing behaviors. In contrast to a simplistic segmentation approach (with two segments) adopted by Chen (2001) and Krishnan and Lecourbe (2010), our approach is a better reflection of the current marketplace and can lead to a firm pricing his products such that complete market coverage is not always optimal. Second, within each market segment, we explicitly consider differences in consumer valuations for both “brown” and “environmental” qualities built into the product offering. Coupled with the market segments incorporated in our paper, this allows us to more clearly specify differences in the willingnessto-pay for each consumer in each segment. In essence, our framework allows us to incorporate the fact that traditional consumers value brown attributes more than environmental attributes; green consumers value environmental attributes more than brown attributes; all the while the consumers in the middle of these two extremes value both attributes equally. Finally, within each strategy choice, we allow for supply side differences in marginal production and fixed costs. This enables strategy choices to be moderated not only by market prices and market coverage but also differences in strategy-specific costs for each product offering.

3 Model preliminaries Our focus here is on a profit-maximizing firm that operates in a consumer market divided into three segments based on customers’ environmental attitudes and shopping behaviors. From a product perspective, consumers have higher expectations for the products they buy and environmental attributes emerge as a new dimension of product quality. Therefore, we posit that a product’s design attributes can be classified into two distinct aggregate dimen-

123

Ann Oper Res

sions: environmental and brown attributes—so that designing a product involves setting each attribute level. An environmental attribute may represent a set of product characteristics that are more environmentally sustainable such as recycled content, energy- and fuel-efficiency, and non-toxicity, while a brown attribute can be considered as a composite measure of safety, availability and style. Both brown and environmental attributes are characterized by quality levels built into the product, and hence consumers always prefer more of attributes than less. From now on, these brown and environmental qualities will be referred to as qb and qe , respectively, leading to the following specific assumption in our paper. Assumption 1 A product is characterized by two quality levels qb and qe , with higher values representing better quality. 3.1 Market structure Customers increasingly incorporate green sentiments into their buying decisions. However, not every customer responds to green products in the same way. To zero in on these customers and so extract the most value out of their products, firms divide their customer market into distinct segments based on customers’ shopping behavior and commitment to green. GfK Roper Public Affairs and Media consistently surveys customers to identify segments based on their environmental attitudes. The results of their recent surveys—conducted in 2002, 2007 and 2008—lead us to define three distinct market segments which are of relevance in examining our product offering strategies:4 (a) Traditionals segment (T): This segment includes customers who are characterized by: (1) being “unconcerned about the environment”; (2) “not caring about the environmental issues and being convinced that buying green products cannot make a difference in helping to protect and improve the environment”; and (3) being “more likely to think that the environmental movement is a front for political interest groups.” The percentage of consumers in this group has steadily declined over time from 37 % in 2002 to 18 % in 2007 to 12 % in 2008. (b) Fence-Sitters segment (F): This segment includes customers characterized by: (1) showing “middling levels of concern about environmental problems” and the fact that “even though they are inclined to adjust their shopping behaviors, they are not utterly resolute about which side of the environmental fence they are on, and so capable of going either way”; (2) having “strong attitudes which have yet to turn to action except for easier behaviors such as recycling”; and (3) having “a lesser concern for the environment than those of the entire population.” The percentage of consumers in this segment has been fairly steady at 46 % in 2002, 41 % in 2007, and 42 % in 2008. (c) Greens segment (G): This segment includes customers who are characterized by: (1) those who are “environmental activists”; (2) those who would always choose green products if they had enough information; and (3) those that are not politically active at this time but as committed to green buying as the activists. The percentage of consumers in this segment has steadily increased over time from 15 % in 2002 to 40 % in 2007 to 46 % in 2008. Based on this segmentation scheme, we define r T , r F , and r G as the proportion of consumers in the Traditionals, Fence-Sitters, and Greens segments, respectively, with 4 The detailed results of the 2002 survey are available in Russo (2008); the detailed results of the 2007 survey

are available at http://ecomerge.blogspot.com/2008/06/gfk-ropers-green-gauge.html (accessed on May 16, 2014); and the detailed results of the 2008 survey are available at http://www.reuters.com/article/2009/03/10/ idUS146105+10-Mar-2009+PRN20090310 (accessed on May 16, 2014).

123

Ann Oper Res

r T + r F + r G = 1. A final key issue relating to market segments is that our approach to analytically characterize the product design and pricing decisions under each strategy choice is based on a two-period analysis. The two aspects which drive our choice of a two-period analysis are: (1) industry data indicate that proportion of customers in each market segment change over time; and (2) it is likely that overall market size would also vary over time. In order to incorporate both of these features, we define M1 and M2 as the total market size for period 1 and period 2, respectively. If r T is the proportion of customers in the Traditionals market segment in period 1, then we assume that r T − δ (0 ≤ δ ≤ r T ) is the proportion of customers in the same segment in period 2. Similarly, if r G is the proportion of customers in the Greens market segment in period 1, then this would translate into r G + δ in period 2, representing the proportion of customers in the same segment in period 2. It is important to note that to be consistent with the changes in distinct market segment sizes observed in the GfK Roper surveys, the proportion of customers in the Fence-Sitters segment is assumed to be unchanged over the two periods. Based on this discussion, the specific market structure assumption is as follows. Assumption 2 r T , r F and r G represent the proportion of consumers in each market segment (0 ≤ r T , r F , r G ≤ 1 such that r T + r F + r G = 1) in period 1, with the total number of consumers in period 1 being M1 . In period 2, the total number of consumers is M2 , with r T − δ, r F and r G + δ representing the proportion of consumers in each market segment (δ ≥ 0 such that 0 ≤ r T − δ, r F , r G + δ ≤ 1). 3.2 Consumer characteristics It is assumed that customers in each market segment are homogeneous in terms of their willingness-to-pay for a product offering. On the other hand, customers across different market segments differ on the ranking of product quality and in their willingness-to-pay. Based on the literature on product differentiation, we use the notion of consumers’ willingness-topay to describe the utility derived by a consumer from a product purchase, and assume that customers’ willingness-to-pay for a product is a linear function of its quality levels (see, for example, Chen 2001 and Krishnan and Lecourbe 2010, among others). Define vb (ve ) as the marginal willingness-to-pay for brown (environmental) quality; p as the price of the product, and 1 − θ (θ ) as the responsiveness of Traditionals (Greens) to environmental (brown) attributes integrated into the product design, where θ ∈ (0, 1). Therefore, the net utility derived by a customer in market segment i (i = T, F, G) from the purchase of a product (qb , qe ) offered at price p is: ⎧ ⎨ vb qb + (1 − θ )ve qe − p, if the customer is in Traditionals segment, if the customer is in Fence-Sitters segment, Ui = vb qb + ve qe − p, (1) ⎩ θ vb qb + ve qe − p, if the customer is in Greens segment. As is obvious, the parameter θ represents the orthogonal adjustments to the marginal valuation of brown and environmental qualities for consumers in Traditionals and Greens segments, respectively. The rationale for this adjustment stems from the following observations. For customers in Greens market segment, the utility function is indicative of their focus on a product’s non-environmental attributes. It can be argued that there does not exist a truly green product as every product uses resources and creates waste. For instance, even though CFLs are energy-efficient, all compact fluorescents contain a tiny yet highly toxic amount of mercury. Conventional alkaline batteries may contain no added mercury, but they are highly toxic nevertheless, because of the other materials they contain. In other times, what may

123

Ann Oper Res

appear to be an environmental benefit may actually lead to non-green product attributes that may at least partially offset benefits/savings to the environment. Hybrid cars, for instance, are fuel-efficient and save energy but the electric batteries they contain represent a potentially significant source of hazardous waste. In these circumstances, Greens segment customers, seeking out products that synchronize with their discerning environmental standards, favor the product based on its green benefits, but may point out the shortcomings of the product’s non-environmental attributes and this can be expressed with a lower value of θ . Conversely, for customers in Traditionals market segment, the adjustment in marginal valuation towards environmental attributes is represented by (1 − θ ). Environmental benefits, such as energy-efficiency and water-efficiency, help cost-conscious Traditionals segment customers save money on energy and water bills over the long term, and so can be a powerful motivator. For example, Energy Star rated washers use at least 50 % less energy and 30 % less water than do traditional washers, while Energy Star refrigerators use at least 15 % less energy than do standard models. Such green products as energy-efficient appliances thus appeal to Traditionals segment consumers’ self-interest while at the same time promoting environmental benefits—a dual feature which would be reflected in a lower θ value. The enforced orthogonality in the value of θ across Traditionals and Greens segments is based on the following rationale. Given that θ represents the focus of a consumer in Greens segment on non-environmental attributes, the GfK Roper surveys note that these two market segments are fairly orthogonal in their thoughts (e.g., Traditionals are “unconcerned about the environment” while Greens are “environmental activists”). Therefore, it seems reasonable to assume that both segment valuations for the attribute not of primary importance would be valued in an orthogonal manner.5 A final point is regarding the utility function of customers in the Fence-Sitters segment. Recall that these customers have environmental attitudes that can cut both ways and so they are capable of going either way—in contrast to customers in the Traditionals and Greens segments who are more fixed in their preferences. Given the Fence-Sitters’ ambivalent attitudes towards buying green products, and the fact that they “can be persuaded to buy green if appealed appropriately” (see Ginsberg and Bloom 2004), the utility function of these customers proposes that shoppers positioned on the “fence” use both brown and environmental attributes as part of their personal valuation of the product offering, and hence fully integrate two distinct qualities into their actual buying decisions. Based on this discussion, our specific assumption on the consumer choice is as follows. Assumption 3 Consumers within each market segment are homogeneous, whereas across market segments, they are heterogeneous in their willingness-to-pay for a product offering. In line with prior research on vertical differentiation models, the net utility a consumer derives from a product purchase is linearly increasing in the quality levels and decreasing in price. 3.3 Strategy alternatives Given the market structure and customer choice characterization, in this paper we examine three mutually exclusive single-product strategies: 5 It could be contended that if we assume symmetric adjustments in the consumer valuations (i.e., if we let θ

represent both Traditionals’ and Greens’ adjustment to the marginal valuations on environmental and brown quality levels, respectively), then our results would be completely different. In an earlier version of this paper, we carried out such an analysis and found that (as would be expected) the exact expressions for Greening-Out and Greening-Up strategies change especially when the firm chooses to price to cover Greens market segment. However, the qualitative insights into strategy choice do not change substantively, and hence we do not present the detailed results for a symmetric adjustment to consumer valuations in this paper.

123

Ann Oper Res

1. Greening-Off : This strategy is characterized by the firm offering a product defined by a single “brown” attribute defined in terms of the quality qb1 ; 2. Greening-Out: Under this strategy, the firm offers a new product that incorporates only environmentally–friendly features, and so the product offering is defined in terms of a single quality-type attribute environmental qe2 ; and 3. Greening-Up: Under this strategy, the firm redesigns his current “brown” product offering to incorporate environmentally–friendly features so that the product offering is defined in terms of two quality-type attributes (qb3 , qe3 ). In essence, if we view these strategies on a continuum, we have the Greening-Off strategy at one end; the Greening-Out strategy at the other end; and the Greening-Up strategy somewhere in the middle. Before discussing the resulting market coverage for each strategy choice, we make an explicit assumption on product pricing that a product is offered at the same price to all consumers in the market. Hence, we do not consider price discrimination across market segments. Our assumption on a common price for each product is of course in line with that of prior work on product design in a vertical differentiation and market coverage framework. However, more importantly it seems to reflect current practice where firms offering green (or partially green) products offer them at the same price to all consumers regardless of the consumer orientation. This leads to the following specific assumption on product pricing associated with each strategy choice. Assumption 4 Under each strategy choice, the firm offers a single product at a common price to all consumers in the market. Given this assumption, the firm has several options to determine the market segments to be covered and this is governed by his product pricing decision. That is, a specific product price could lead to either all consumers in a specific segment deriving a non-negative net utility and so choosing to purchase the product; or all consumers in the same segment deriving a negative net utility so that none of them would purchase the product. Examining the utility functions under each strategy alternative, the following pricing options are available to the firm: 1. Greening-Off : Given that the firm offers a single “brown” product (qb1 , p) under this strategy, the net utility of all customers in each market segment are: UT = vb qb1 − p U F = vb qb1 − p UG = θ vb qb1 − p

(2)

Since θ ∈ (0, 1), the firm has two distinct pricing options: i. Setting p = vb qb1 , and hence covering market segments T and F completely; and ii. Setting p = θ vb qb1 , and covering all market segments (i.e., T , F and G) completely. 2. Greening-Out: Given that the firm offers a single “green” product (qe2 , p) under this strategy, the net utility of all customers in each market segment are: UT = (1 − θ ) ve qe2 − p U F = ve qe2 − p UG = ve qe2 − p

(3)

Since θ ∈ (0, 1), the two distinct pricing options are:

123

Ann Oper Res

i. Setting p = ve qe2 , and hence covering market segments F and G completely; and ii. Setting p = (1 − θ ) ve qe2 , and covering all market segments (i.e., T , F and G) completely. 3. Greening-Up: Because the firm offers a single “greened-up” product (qb3 , qe3 , p) under this strategy, the net utility of all customers in each market segment are: UT = vb qb3 + (1 − θ ) ve qe3 − p U F = vb qb3 + ve qe3 − p UG = θ vb qb3 + ve qe3 − p

(4)

Given θ ∈ (0, 1), there are three distinct pricing options available under this strategy choice: i. Setting p = vb qb3 + ve qe3 , and hence covering market segment F completely; ii. Setting p = θ vb qb3 + ve qe3 , and completely covering all market segments (i.e., T , F and G) if vb qb3 / (vb qb3 + ve qe3 ) ≥ θ , and only segments F and G if vb qb3 / (vb qb3 + ve qe3 ) < θ ; and iii. Setting p = vb qb3 + (1 − θ ) ve qe3 , and completely covering all market segments (i.e., T , F and G) if vb qb3 / (vb qb3 + ve qe3 ) ≤ θ , and only segments T and F if vb qb3 / (vb qb3 + ve qe3 ) > θ . Although it might be argued that only those pricing options which result in complete coverage of all segments should be the dominant choices, this is not always the case because the supply parameters would also be relevant in making the optimal pricing decision. These supply parameters for each strategy choice are described in the following section. 3.4 Supply parameters On the supply side, the firm attempts to serve all the customers in targeted market segment(s) with either a single-attribute product (i.e., a brown product represented solely by brown quality qb1 or a green product represented solely by environmental quality qe2 ) or a two-attribute product (i.e., a greened-up product composed of both qualities qb3 and qe3 ). Based on the argument put forward by Moorthy (1984), we also assume that unit production cost increases at a faster rate than consumers’ willingness-to-pay, and the cost of supplying a product— regardless of whether it is a brown, green or greened-up product—increases quadratically with its quality level(s). More specifically, we assume that the unit cost of supplying a brown 2 and c q 2 , respectively, where product (qb1 ) and a green product (qe2 ) is given by cb qb1 e e2 cb and ce are positive cost coefficients, and the unit cost of production for the greened-up product (qb3 , qe3 ) is given by cu (qb3 + qe3 )2 with cu > 0. The cost coefficients cb , ce and cu reflect the differences in costs of producing a product across different quality types. In addition, although there could be learning effects from the supply side, we adopt a worst-case scenario and assume that per-unit variable production costs are time-invariant. In addition to the per-unit variable production costs, the total cost of developing a green product and a greened-up product comprises another strategy specific components. In implementing the Greening-Out strategy, the firm incurs up-front costs from introducing a new green product into the market for process R&D, market research and other relevant expenses. Drawing upon the literature on new product development (see, for example, Chen 2001), the presumed sunk costs of developing and launching the green product is represented by a constant N > 0. Under the strategy of Greening-Up, the firm improves the environmental quality of his existing brown product by designing bad materials out of the product or replacing its

123

Ann Oper Res

bad materials with better ones. Such R&D related efforts of the firm directly translate into a fixed cost of product redesign (e.g., cost of developing renewable packaging material), which is assumed to be convex and strictly increasing in the environmental quality qe3 designed 2 with F > 0.6 into the existing brown product, and given by Fqe3 4 Model analysis In a two-period setting, the firm’s key decision is to specify the quality level(s) to be supplied under a strategy choice at time 0. Before structurally analyzing each strategy choice, we make one additional assumption. Note that our market structure characterization in Sect. 3.1 allows for the proportional change in consumers between segments between two time periods. Our analysis is, however, restricted to the firm considering a single strategy choice over the two periods. The basis for this assumption is threefold. First, it allows us to identify dominant single-strategy choices when the firm has a priori information on evolving market structures which is the focus of our paper. This in turn enables us to explore the trade-offs between partial versus complete market coverage in making the pricing decisions under each strategy choice. Second, it is in line with the focus of our work since it provides guidelines for firms interested in introducing a single product rather than a product combination to the market. Third, it provides a foundation for exploring issues related to introducing combinations of products rather than a single product which in itself is an important first step in analyzing green product introduction decisions. The explicit assumption stemming from these arguments is as follows. Assumption 5 In the analysis that follows, we evaluate three mutually exclusive choices for the firm: Greening-Off; Greening-Out; and Greening-Up. To appropriately account for the realized firm-level profits under each strategy choice, we define e1 and e2 as the continuous discount factors for periods 1 and 2, respectively. The complete notation for our paper is defined in Table 1. 4.1 Greening-Off strategy Under this strategy, the firm offers a single brown product with quality level qb1 without any concern for environmental considerations. The firm’s potential shopper base is made up of Traditionals, Fence-Sitters and Greens segments, and depending upon his pricing decision, the brown product can potentially be attractive to consumers in each segment. Recall that a brown product of quality qb1 delivers a value of vb qb1 to consumers in Traditionals and Fence-Sitters segments, while Greens value the same product differently at θ vb qb1 with 0 < θ < 1. Therefore, there are two potential targeting strategies available to the firm: i. The firm can target his existing brown product specifically at only Traditionals and FenceSitters consumers by setting a selling price equivalent to these two segments’ willingnessto-pay for a unit of traditional quality (i.e., p = vb qb1 )—a move that enables the firm to extract all consumer surplus from Traditionals and Fence-Sitters segments. Nonetheless, asking Greens to accept a product without any environmental benefits at fairly high prices is likely to keep these customers away from purchasing the product because the valuation 6 The quadratic structure of product redesign cost is an approximation of a general convex cost function

widely observed in practice, and it has been commonly employed in the literature (see, for example, Lacourbe et al. 2009, among others).

123

Ann Oper Res Table 1 Notation Decision variables qb

Traditional (brown) quality of a product

qe

Environmental (green) quality of a product

p

Price of a product

Market structure rT , r F , rG

Proportion of consumers in Traditionals, Fence-Sitters and Greens segments

M1 , M2

Total market size for periods 1 and 2

δ

Decrease (increase) in proportion of consumers in Traditionals (Greens) segment

Consumer characteristics vb

Marginal willingness-to-pay on brown quality

ve

Marginal willingness-to-pay on environmental quality

Θ

Responsiveness of Greens to brown quality

Costs cb , ce , cu

Variable cost coefficients of brown, green and greened-up products

F

Product redesign cost of a greened-up product

N

Fixed costs of developing and introducing a green product

they place on the brown product would fall below its selling price (i.e., θ vb qb1 < p). Hence, if this pricing option is adopted, the firm’s profit maximization problem gets the following form:   2 max Π1i = (e1 M1 (r T + r F ) + e2 M2 (r T + r F − δ)) vb qb1 − cb qb1 (5) qb1 ≥0

Clearly, this profit function is strictly concave in qb1 , and by setting the first order condition equal to 0, the optimal brown quality level and associated firm profits are obtained as follows: vb 2cb ((r T + r F ) (e1 M1 + e2 M2 ) − e2 M2 δ) vb2 = 4cb

∗ qb1 =

Π1i

∗

(6) (7)

ii. The firm can attempt to cover his entire customer base by jointly addressing customers in all market segments—a strategy that works only if the firm sets a selling price for his brown product low enough such that it matches the moderately low willingness-to-pay of Greens segment customers for brown product attributes (i.e., p = θ vb qb ). This allows the firm to extract the entire consumer surplus from Greens segment, while creating a positive surplus for Traditionals and Fence-Sitters customers. In this setting, the firm’s profit maximization problem is:   2 max Π1ii = (e1 M1 + e2 M2 ) θ vb qb1 − cb qb1 (8) qb1 ≥0

It is apparent that this profit function is strictly concave in qb1 , and by setting the first order condition equal to 0, the optimal brown quality level and corresponding firm profits are derived as follows:

123

Ann Oper Res

θ vb 2cb (e1 M1 + e2 M2 ) θ 2 vb2 = 4cb

∗ qb1 =

Π1ii

∗

(9) (10)

Based on this analysis, it is possible to structurally characterize conditions under which the firm should choose between these two distinct pricing options and the proposition below provides this condition. Proposition 1 Given θ ∈ (0, 1), under the Greening-Off strategy, the firm should choose to serve all market segments if and only if θ > θ A where   e1 M1 r G + e2 M2 (r G + δ) . (11) θA = 1 − e 1 M 1 + e2 M 2  

Proof See “Appendix 1”.

It is seen in Proposition 1 that there are two specific aspects which moderate the pricing 2 M2 (r G +δ) decision for the firm. Note that e1 M1er1GM+e can be defined as the weighted propor1 +e2 M2 tion of market sizes across both periods for Greens market segment, while θ is simply the discounted marginal valuation by consumers in the same segment. In essence, the proposition states that pricing to capture the entire market is an optimal strategy either when: (a) the weighted proportion of market sizes is small and the marginal valuation of consumers in Greens segment is large; or (b) the weighted proportion of market sizes is large and the marginal valuation of consumers in Greens segment is small. Market coverage strategies with respect to θ A under this strategy is illustrated in Fig. 1. 1

Complete Coverage

θA

θ Traditionals & Fence-Sitters

rG

0

1

rT

Fig. 1 Optimal market coverage and threshold θ A under Greening-Off strategy

123

Ann Oper Res

4.2 Greening-Out strategy Under this strategy, driven by the underlying fact that the burgeoning and highly desirable Greens segment shoppers have unmet needs and wants that are attributed to their environmental concerns and that await a solution, the brown firm discontinues selling his brown product and embarks upon an entirely new or different product concept: a new green product built from scratch with quality level qe2 , representing minimal (if not zero) environmental impact. The primary end goal of this radical change is to get an environmentally preferable substitute for the traditional brown product into the hands of Greens and Fence-Sitters consumers who place a high priority on environmental quality. Nevertheless, the firm can also improve consumer appeal for his environmentally-sensitive product by positioning its environmental features into advantages that resonate with Traditionals segment consumer values. When value added benefits of an environmentally–friendly product (e.g., cost-savings) appeal to Traditionals segment consumers, they are more amenable to buying it. The two distinct pricing strategies that enable the firm to capture the full possible value of his new green offering by strongly influencing consumer purchase decisions can be described as follows: i. The firm can vary his green product’s price according to Greens and Fence-Sitters segment customers’ latitude of price acceptance, which is a range of possible prices within which price changes have little or no impact on these customers’ purchase decisions, expressed by ve qe2 . By pricing in this manner, the firm may very well raise his profit margins 2 ) but suffer from a guaranteed decrease in its sales due to per product to (ve qe2 − ce qe2 exclusion of Traditionals from the consumption activity. In this case, the firm’s profit maximization problem would be   2 max Π2i = (e1 M1 (r G + r F ) + e2 M2 (r G + r F + δ)) ve qe2 − ce qe2 − N . (12) qe2 ≥0

This profit function is apparently strictly concave in qe2 , and by setting the first order condition equal to 0, the optimal environmental quality level and associated firm profits are obtained as follows: ve ∗ qe2 = (13) 2ce ((e1 M1 + e2 M2 ) (r G + r F ) + e2 M2 δ) ve2 ∗ −N (14) Π2i = 4ce ii. In a continuing effort to increase the collective market share of his new green product through additionally capturing Traditionals segment customers, the firm can push the sales of his green product by adjusting its prices according to Traditionals’ willingnessto-pay for the environmental benefits it offers (i.e., p = (1 − θ ) ve qe2 ). Pricing based on Traditionals’ valuation on green product not only entices these customers to purchase the green product they otherwise would not have considered, but also creates a positive surplus for Greens and Fence-Sitters who are willing to pay higher prices and so who will benefit from lower prices. This increases the number of customers that could be served under the previous pricing option. Based on that, the firm’s profit maximization problem would be   2 max Π2ii = (e1 M1 + e2 M2 ) (1 − θ ) ve qe2 − ce qe2 − N. (15) qe2 ≥0

Clearly, this profit function is strictly concave in qe2 , and on the basis of the first order condition, the optimal green quality level and corresponding firm profits are derived as follows:

123

Ann Oper Res

(1 − θ ) ve 2ce (e1 M1 + e2 M2 ) (1 − θ )2 ve2 = −N 4ce ∗ qe2 =

Π2ii

∗

(16) (17)

Now we can compare these two distinct pricing options the firm can employ under the Greening-Out strategy to transform his new green product into a profitable offering, and identify what factors are driving the firm’s choice of each. Proposition 2 Given θ ∈ (0, 1), under the Greening-Out strategy, the firm would be better off serving all market segments if and only if θ < θ B where   e1 M1 r T + e2 M2 (r T − δ) . (18) θB = 1 − 1 − e 1 M 1 + e2 M 2 Proof See “Appendix 2”.

 

Mirroring the findings presented in Proposition 1, Proposition 2 also shows that there are two specific aspects which moderate the pricing decision of the firm. Note that e1 M1 r T +e2 M2 (r T −δ) can be defined as the weighted proportion of market sizes across both e1 M1 +e2 M2

periods for Traditionals market segment, while θ is simply the discounted marginal valuation by consumers in the same segment. Hence, pricing to capture the entire market is an optimal strategy either when: (a) the weighted proportion of market sizes is small and the marginal valuation of consumers in Traditionals segment is large; or (b) the weighted proportion of market sizes is large and the marginal valuation of consumers in Traditionals segment is small. Market coverage strategies with respect to θ B under this strategy is illustrated in Fig. 2. 4.3 Greening-Up strategy Under this strategy, the firm’s existing brown product is streamlined to reduce waste, cut energy and water use, lower carbon-emissions, minimize use of materials, and so forth. In other words, the firm now offers a product with both brown and environmental qualities, represented by (qb3 , qe3 ). Note that the additional environmental benefit(s), properly connected with tastes and buying behavior of shoppers with differing environmental attitudes, help these customers to better appreciate the prevailing product. This in turn merits the firm’s asking for higher prices for additional environmental features to cover the added costs over his existing brown product. In doing so, the firm can extract additional economic value by redesigning his existing brown product. Because overall customer market is comprised of Traditionals, Fence-Sitters and Greens segments, the firm can choose among three different pricing options to better leverage the heterogeneity in valuation of customers in different segments. These pricing options are as follows: i. The firm can equate the price of his greened-up product (qb3 , qe3 ) with Fence-Sitters segment customers’ willingness-to-pay (i.e., p = vb qb3 + ve qe3 ). However, offering the same brown product with additional environmental benefits at such a price would lead to loss of customers served in Traditionals and Greens segments, since these customers’ willingness-to-pay would be lower than the price. In fact, these customers would resist the “overpriced” product. Therefore, under this pricing option, the profit maximization problem for the firm becomes   2 max Π3i = r F (e1 M1 + e2 M2 ) vb qb3 + ve qe3 − cu (qb3 + qe3 )2 − Fqe3 . (19) qb3 ;qe3 ≥0

123

Ann Oper Res 1

Greens & Fence-Sitters

θ θB

Complete Coverage

rG

0

1

rT

Fig. 2 Optimal market coverage and threshold θ B under Greening-Out strategy

This profit function is strictly and jointly concave in qb3 and qe3 , and hence by setting the first order conditions equal to 0, we can determine the optimal quality decisions and corresponding profits for the firm as follows:7 vb r F (e1 M1 + e2 M2 ) (ve − vb ) − 2cu 2F M + e M r (e F 1 1 2 2 ) (ve − vb ) ∗ = qe3 2F 

vb2 r F (e1 M1 + e2 M2 ) (ve − vb )2 = r F (e1 M1 + e2 M2 ) + 4cu 4F ∗ qb3 =

Π3i

∗

(20) (21) (22)

ii. Instead of appending environmental benefits at raised prices that are balanced against Fence-Sitters’ willingness-to-pay, the firm can charge what the combination of brown and environmental attributes designed into the greened-up product is worth to Greens segment customers who are loath to pay high prices for non-green attributes (i.e., p = θ vb qb3 + ve qe3 ). From the Fence-Sitters’ perspective, such a pricing approach should mean that perceived benefits for Fence-Sitters from purchasing the greened-up product would outweigh its costs, and they would be willing (even eager) to buy the greened-up product. That’s said, the product purchase decision of Traditionals cannot ∗ be simply determined because it is dependent on the optimal quality levels (i.e., qb3 ∗ ). More notably, to determine these optimal quality levels, we need to know a and qe3 priori whether the Traditionals do or do not purchase the product (i.e., we have “what came before—the chicken or the egg” problem!). To resolve this dilemma, we define: 7 For feasibility (i.e., for both q b3 and qe3 to be non-negative), we assume that 1 ≤ F . cu r F (e1 M1 +e2 M2 )

123

ve vb

≤ 1 +

Ann Oper Res

(a) x1 = e1 M1 + e2 M2 , if Traditional segment customers buy the product; and (b) x2 = (r G + r F ) (e1 M1 + e2 M2 ) + e2 M2 δ, if they choose not to do so. Thus, the profit maximization problem for the firm under this pricing option becomes   2 max Π3ii = xˆ θ vb qb3 + ve qe3 − cu (qb3 + qe3 )2 − Fqe3 , (23) qb3 ;qe3 ≥0

where xˆ can be set equal to either x1 or x2 as defined earlier. Regardless of the value of x, ˆ this profit function is strictly and jointly concave in qb3 and qe3 , and so the optimal quality levels and corresponding profits for the firm can be determined by means of the first order conditions as follows:8 θ vb xˆ (ve − θ vb ) − 2cu 2F − θ vb ) x ˆ (v e ∗ = qe3 2F

 θ 2 vb2 x(v ˆ e − θ vb )2 = xˆ + 4cu 4F

∗ qb3 =

Π3ii

∗

(24) (25) (26)

Examining the solution to this problem, the optimal market coverage for this pricing option under Greening-Up strategy is given in the following proposition. Proposition 3 Given θ ∈ (0, 1) and cu > 0, when greened-up product is priced based on Traditionals segment customers’ willingness-to-pay, the optimal market coverage is: a. If 1 ≤

ve F θ vb ≤ 1 + cu x1 , then all market segments are covered; and ve F F cu x1 < θ vb ≤ 1 + cu x2 , then only Fence-Sitters and Greens

b. If 1 + covered,

market segments are

where x1 = e1 M1 + e2 M2 and x2 = (r G + r F ) (e1 M1 + e2 M2 ) + e2 M2 δ.  

Proof See “Appendix 3”.

iii. In contrast to basing the price on Greens’ willingness-to-pay, the price tag of the firm’s greened-up product can reflect Traditionals segment customers’ valuation for the brown and environmental benefits the “refreshed” brown product offers (i.e., p = vb qb3 + (1 − θ ) ve qe3 ). In this setting, it is the coverage of Greens segment which is uncertain and based on a reasoning similar to that detailed in the preceding case, the firm’s profit maximization problem is   2 max Π3iii = yˆ vb qb3 + (1 − θ ) ve qe3 − cu (qb3 + qe3 )2 − Fqe3 , (27) qb3 ;qe3 ≥0

where yˆ is set equal to either x1 (as defined earlier) or x3 = (r T + r F )(e1 M1 + e2 M2 ) − e2 M2 δ. It is relatively easy to show that this profit function is strictly and jointly concave in qb3 and qe3 , and so the optimal quality decisions and corresponding profits for the firm can be derived by setting the first order conditions equal to 0 as follows:9 8 For feasibility (i.e., for both q and q to be non-negative), we assume that 1 ≤ ve ≤ 1 + F . e3 b3 cu x2 θvb 9 For feasibility (i.e., for both q and q to be non-negative), we assume that 1 ≤ (1−θ)ve ≤ 1 + F . e3 b3 vb cu x3

123

Ann Oper Res

vb yˆ ((1 − θ ) ve − vb ) − 2cu 2F yˆ ((1 − θ ) ve − vb ) ∗ = qe3 2F

 vb2 yˆ ((1 − θ ) ve − vb )2 = yˆ + 4cu 4F

∗ qb3 =

Π3iii

∗

(28) (29) (30)

Examining the solution to this problem, the optimal market coverage for this pricing option under Greening-Up strategy is given in the following proposition. Proposition 4 Given θ ∈ (0, 1) and cu > 0, when greened-up product is priced based on Greens segment customers’ willingness-to-pay, the optimal market coverage is: (1−θ )ve ≤ 1 + cuFx1 , vb )ve ≤ + cuFx1 < (1−θ vb

a. If 1 ≤

then all market segments are covered; and

b. If 1 segments are covered,

1+

F cu x 3 ,

then only Traditionals and Fence-Sitters market

where x1 = e1 M1 + e2 M2 and x3 = (r T + r F )(e1 M1 + e2 M2 ) − e2 M2 δ. Proof See “Appendix 4”.

 

Based on the arguments and findings provided above, the following proposition outlines the optimal pricing decisions of the firm under Greening-Up strategy. Proposition 5 Given θ ∈ (0, 1) and cu > 0, the firm’s optimal pricing decision under Greening-Up strategy choice is as follows: a. When θ ≤ ve /vb < 1, it is optimal to set p = θ vb qb3 + ve qe3 (i.e., case ii); b. When 1 ≤ ve /vb < 1/ (1 − θ ), it is optimal to set  p = vb qb3 + ve qe3 (i.e., case i) if cu (e1 M1 + e2 M2 ) (ve − θ vb )2 − r F2 (ve − vb )2 − Fvb2 r F − θ 2 < 0, and p = θ vb qb3 + ve qe3 (i.e., case ii), otherwise; and c. When ve /vb ≥ 1/(1 − θ ), it is optimal to set p = θ vb qb3 + ve qe3 (i.e., case ii) if cu (e1 M1 + e2 M2 ) (ve − θ vb )2 − ((1 − θ ) ve − vb )2 − Fvb2 1 − θ 2 > 0, and p = vb qb3 + (1 − θ ) ve qe3 (i.e., case iii), otherwise. Proof See “Appendix 5”.

 

The insights stemming from these results are that in all cases pricing to capture Greens segment customers is always a choice. It is only in certain extreme situations that either one of the other two pricing options emerges as a dominant choice. Note also that the feasibility of each pricing option under Greening-Up strategy depends upon the relative values of ve /vb and θ . For example, when ve /vb is greater than or equal to θ but strictly less than 1, the only feasible option (and so the optimal pricing strategy) is to set the price of the greened-up product equal to Greens segment customers’ willingness-to-pay. On the other hand, when ve /vb ≥ 1/(1 − θ ), the firm has to make a choice between pricing on the basis of Greens and Traditionals customers’ product valuation. In this case, additional product- and marketrelated factors (i.e., cu , F, e1 M1 and e2 M2 ) enter into the equation to determine the optimal pricing regime under Greening-Up strategy. Our focus now shifts to evaluating and comparing three strategy choices with different pricing options.

123

Ann Oper Res

5 Strategy comparisons and insights In this section, we focus on economic and environmental issues germane to greener products, and examine how the characteristics of green market environment (i.e., relative sizes of green consumer segments and consumers’ product valuations) create incentives for a firm to adopt a specific green product strategy. In doing so, we encounter a major hurdle in that an analytical comparison of strategies seems to be very difficult due to multiple strategy- and productspecific parameters, and alternative pricing options under each strategy choice. That is said, it is important to note that existent consumer choice data provide valuable insights into the values of relevant market parameters used in our analysis. To start with, we set the relative size of each market segment in period 1 equal to the respective ratios specified in 2008 GfK Roper market survey: r T = 0.12, r F = 0.42 and r G = 0.46. For period 2, the reduction δ (increase) in proportion of customers in Traditionals segment (Greens segment) is set equal to 0.04. This value is derived by averaging out the decline in that proportion from 2002 to 2008 as indicated by the GfK Roper surveys. We normalize the discounted market size in the first period to 1 (i.e., e1 M1 = 1). For the second period, we set e2 M2 = 1.07 to capture the findings of 2008 GfK Roper survey reports that there is a 7 % increase in the number of customers who think more about “green” from 2007 to 2008. On the supply side, from a product-attribute perspective, previous literature on product design and green product development (see, for example, Chen 2001 and Liu et al. 2012, among others) attribute: (a) the lowest coefficient of unit production cost to products with only brown attributes; (b) the highest coefficient of unit production cost to products with only environmental attributes; and (c) the moderate coefficient of unit production cost to products defined by a combination of both brown and environmental attributes. We draw upon these studies and set the per product unit cost coefficients cb , ce and cu at $1.0, $2.0 and $1.5, respectively. On the basis of these market and supply parameters, we start by considering two extreme cases by considering θ = 0 and θ = 1. When θ equals 0, it means that Greens segment customers do not derive any additional value from the non-environmental attributes of a product. On the other hand, because 1 − θ gets equal to 1 when θ = 0, it also implies that environmental attributes of the product completely resonate with Traditionals segment customer values. In other words, the environmental attributes of the product offer “non-green” consumer value (or mainstream appeal) such as cost savings, convenience and health. As a case in point, the appeal of CFL bulbs can be attributed to their energy savings and long life-qualities that make them convenient and economical over time. As shown by market evidence, when these customers are convinced of such desirable “non-green” benefits of environmental products, they are more inclined to adopt them. When θ = 0, Traditionals segment customers no longer recognize the environmental product as green because they buy it for non-green reasons. In this context, the dominance relationships between strategies can be characterized as presented in the following proposition and in Fig. 3. Proposition 6 Given that F > 0 and N > 0, when θ = 0: – If ve /vb ≤ 1, then – Greening-Off with pricing to cover T and F is the dominant strategy if ve /vb < √ 0.78 + 0.25F, and

123

Ann Oper Res

Fig. 3 Strategy choices and market coverage when θ = 0

– Greening-Up with pricing to cover either market √ segments F and G or all market segments is the dominant strategy if ve /vb ≥ 0.78 + 0.25F; – If ve /vb > 1 and N is prohibitively high, then – Greening-Off √ with pricing to cover market segments T and F is the dominant strategy if ve /vb ≤ 0.75 + 0.24F, and – Greening-Up with pricing to cover either market √ segments F and G or all market segments is the dominant strategy if ve /vb > 0.75 + 0.24F. Proof See “Appendix 6”.

 

When θ equals 1, because 1 − θ gets equal to 0, environmental benefits of a product appeal to only Greens and Fence-Sitters customers. These attributes do not fulfill the needs and interests of Traditionals segment customers beyond what is good for the environment. Traditionals segment customers are outside the green niche and they are not attracted into purchasing the greener product unless the product does offer such a desirable benefit as cost effectiveness, health and safety. Take inkjet cartridges, for example. They can be made with post-consumer plastics by using a process that combines various kinds of recycled plastics, from used inkjet cartridges to common water bottles, with a variety of chemical additives. This can save the energy needed to ship and produce new materials, preventing in turn the air pollution associated with the generation of new energy. Even though its recycled content makes the product better for the environment and so makes it appealing to Greens segment customers, this product is quite unlikely to appeal to Traditionals segment customers’ selfinterest or needs. In such cases, θ = 1 indicates that Traditionals are indifferent to the value offered by the environmental benefits of a greener product and so they do not derive additional value from it, whereas Greens completely approve the non-environmental attributes of the product. In this context, the dominance relationships between strategies can be characterized as presented in the following proposition and in Fig. 4.

123

Ann Oper Res

Greening-Off

Greening-Off

Greening-Up

T &F &G

T &F

F

0

1

ve /vb

√ 1 + 0.49 F Fig. 4 Strategy choices and market coverage when θ = 1

Proposition 7 Given that F > 0 and N > 0, when θ = 1: – If ve /vb ≤ 1, then – Greening-Off with pricing to cover all market segments is always the dominant strategy; – If ve /vb > 1 and N is prohibitively high, then – Greening-Off with pricing to cover market segments T and F is the dominant strategy √ if ve /vb ≤ 1 + 0.49 F, and – Greening-Up with√pricing to cover market segment F is the dominant strategy if ve /vb > 1 + 0.49 F. Proof See “Appendix 7”.

 

We now turn our attention to the more general case where θ ∈ (0, 1). Based on the settings for market-side and supply-side parameters, we show the optimal quality level(s) and firm-level profits for each market coverage option (as determined by the chosen pricing regime) under each strategy choice in Table 2. Notice that the strategy choice of a firm that would maximize his profits is determined by the relative values of parameters ve and vb , and the value of parameter θ , along with the strategy specific cost parameters F and N . Recall from Sect. 3.2 that depending on certain product-related and market-related factors, Greens customers’ sensitivity to non-environmental attributes of a product θ can take any value between 0 and 1. When we merge the feasibility and optimality conditions of distinct pricing regimes under Greening-Up strategy (presented in Proposition 5) with those of Greening-Off and GreeningOut strategies (presented in Propositions 1 and 2, respectively), we are able to obtain Fig. 5 with thresholds θ A and θ B (defined in Propositions 1 and 2, respectively). Note that each designated region in Fig. 5 defines a set of candidate optimal strategies based on the relative values of ve /vb and θ , and the value of θ relative to thresholds θ A and θ B . For example, region D points to the set of strategy-pricing options when ve /vb ∈ [θ, 1] and θ ∈ [0, θ B ]. In each of these regions, the optimal strategy choice and subsequent pricing regime of the firm are dependent upon the cost parameters F and N , specific to Greening-Up and Greening-Out strategies, respectively. The candidate dominant strategies in each region depicted in Fig. 5, along with customer segments to be covered by each strategy, are provided in detail in Table 3. The economic implications of green product strategies reflect that a firm can derive economic benefits from “design for environment” under different market conditions. However, green product strategy choice of a firm cannot be determined by economic criteria alone. Creating bottom-line benefits can be only one part of choosing a green product strategy because firms are under ever-increasing pressure from governments (in addition to customer market and supply chain partners) to reduce their environmental impact. In other words, it is

123

123

T, F

T, F, G

F, G

T, F, G

F

T, F, G

T, F, G

Greening-Off (i)

Greening-Off (ii)

Greening-Out (i)

Greening-Out (ii)

Greening-Up (i)

Greening-Up (ii)

Greening-Up (iii)

(0.23)ve2 − N (0.26) (1 − θ )2 ve2 − N (0.14)vb2 + (0.19)(ve − vb )2 /F (0.35)θ 2 vb2 + (1.07)(ve − θ vb )2 /F (0.35)vb2 + (1.07)((1 − θ ) θ ve − vb )2 /F

(0.25)ve (0.25)(1 − θ ) ve (0.43) (ve − vb )/F (1.04) (ve − θ vb )/F (1.04) ((1 − θ ) ve − vb )/F

NA

(0.33)vb − (0.43) (ve − vb )/F (0.33)θ vb − (1.04) (ve − θ vb )/F

(0.33)vb − (1.04) ((1 − θ ) ve − vb )/F

NA

(0.52)θ 2 vb2

NA

(0.50)θ vb

(0.27)vb2

NA

(0.50)vb

Profit Environmental

Quality levels Brown

For Greening-Up strategy choice, the last two pricing options always lead to complete market coverage (see Propositions 3 and 4)

Market segments

Strategy

Table 2 Optimal quality levels and profits for strategy choices

Ann Oper Res

Ann Oper Res

ve /vb

L

ve /vb = 1/(1 − θ)

K J I H G

1

F E

ve /vb = θ

D

C B A

0

θB

θA

1

θ

Fig. 5 Parameter regions depicting the candidate optimal strategy-pricing options

important to choose a green product strategy that would improve environmental performance of a firm, all the while boosting his top-line revenues. To explore the extent to which a green product strategy creates higher environmental benefits while providing economic payoffs to a firm, we compare the overall environmental quality that can be supplied by a firm under Greening-Out and Greening-Up strategies. Our focus is to address the issue of under what conditions a firm adopts a green product strategy not only because ‘it is the most profitable path to follow’ but also because ‘it is the most environmentally friendly approach to adopt’. Before focusing on the key drivers of meeting a firm’s economic and environmental objectives together, we look at under what conditions one of the green product strategies outgreens the other (or vice versa)—regardless of the economic criteria. For many firms, key competitors are already playing in the green consumer space, making it necessary to look for opportunities to outperform these competitors on greenness or at least match their environmental performance (e.g., see Ginsberg and Bloom 2004). To compare the total environmental quality supplied by a firm under Greening-Out and Greening-Up strategies, we use the optimal environmental quality values presented in Table 2 and take the aggregate sum of environmental quality specified in all new green products or greened-up products in the market, respectively.10 Note that total environmental quality supplied by a firm differs not only between two strategies but also across distinct pricing (and so market coverage) options under each strategy choice as outlined in Table 4. 10 The overall environmental quality is assumed to be linearly additive across all products. See Chen (2001) and Krishnan and Lecourbe (2010).

123

Ann Oper Res Table 3 Regions, candidate optimal strategy choices and market coverage Region

Range for θ

Range for ve /vb

Strategies

Market segments

A

(0, θ B ]

(0, θ ]

Greening-Off (i)

T, F

B

[θ B , θ A ]

(0, θ ]

Greening-Off (i)

T, F

C

[θ A , 1)

(0, θ ]

Greening-Off (ii)

T, F, G

D

(0, θ B ]

[θ, 1]

Greening-Off (i)

T, F

Greening-Up (ii)

T, F, G T, F

E

[θ B , θ A ]

[θ, 1]

Greening-Off (i) Greening-Up (ii)

T, F, G

F

[θ A , 1)

[θ, 1]

Greening-Off (ii)

T, F, G

Greening-Up (ii)

T, F, G

(0, θ B ]

1 1, (1−θ)

Greening-Off (i)

T, F

Greening-Up (i)

F

G

H

I

[θ B , θ A ]

[θ A , 1)

1 1, (1−θ)



1 1, (1−θ)

J

(0, θ B ]

[1/(1 − θ ), −)

K

[θ B , θ A ]

[1/(1 − θ ), −)

L

[θ A , 1)

[1/(1 − θ ), −)

Greening-Up (ii)

T, F, G

Greening-Off (i)

T, F

Greening-Out (i)

F, G

Greening-Up (i)

F

Greening-Up (ii)

T, F, G

Greening-Off (ii)

T, F, G

Greening-Out (i)

F, G

Greening-Up (ii)

T, F, G

Greening-Off (i)

T, F

Greening-Up (ii)

T, F, G

Greening-Off (i)

T, F

Greening-Out (i)

F, G

Greening-Up (ii)

T, F, G

Greening-Up (iii)

T, F, G

Greening-Off (ii)

T, F, G

Greening-Out (i)

F, G

Greening-Up (ii)

T, F, G

Greening-Up (iii)

T, F, G

θ B = 0.05 and θ A = 0.72 based on available consumer choice data and expressions provided in Propositions 1 and 2 Table 4 Total environmental quality under Greening-Out and Greening-Up strategies Strategy

Greening-Out Greening-Up

123

Market coverage

Total environmental quality (TEQ)

F, G

(0.47)ve

T, F, G

(0.52) (1 − θ ) ve

F

(0.37) (ve − vb ) /F

T, F, G

(2.15) (ve − θ vb ) /F

T, F, G

(2.15) ((1 − θ )ve − vb ) /F

Ann Oper Res θ = 0.10

ve /vb

ve /vb =

Greening-Up

4.57θ 4.57−F

1

ve /vb = θ ve /vb =

4.13θ 4.13−F (1−θ)

Greening-Out

0

1

θ

Fig. 6 Environmental performance of Greening-Up versus Greening-Out strategies

The comparison of the total environmental qualities supplied under Greening-Out and Greening-Up strategies (i.e., TEQ2 and TEQ3 , respectively) yields the following result (Fig. 6). Proposition 8 Given θ ∈ (0, 1) and F > 0: – If ve /vb > θ and θ ≤ 0.10, then – Greening-Up outgreens Greening-Out (i.e., TEQ3 > TEQ2 ) if and only if 4.13θ ve > vb 4.13 − F(1 − θ )

(31)

– Otherwise, Greening-Out outgreens Greening-Up (i.e., TEQ2 > TEQ3 ) – If ve /vb > θ and θ > 0.10, then – Greening-Up outgreens Greening-Out (i.e., TEQ3 > TEQ2 ) if and only if 4.57θ ve > vb 4.57 − F

(32)

– Otherwise, Greening-Out outgreens Greening-Up (i.e., TEQ2 > TEQ3 ) – If ve /vb ≤ θ , then – Greening-Out is always the dominant strategy on environmental basis (i.e., TEQ2 > TEQ3 ). Proof See “Appendix 8”.

 

123

Ann Oper Res

The prevailing view on the link between design for environment and environmental impact states that even though developing a green product from scratch enables firms to address environmental concerns in detail in their products/operations, it is the incremental changes made on an existing non-green product that would significantly improve the environmental performance of firms (see Hopkins 2010, for example). On the other hand, Proposition 8 shows that this view is not necessarily true and greening up an existing brown product can be better at reducing the overall environmental impact than designing a new green product if certain product- and market-related conditions hold. In essence, we observe two major constraints on the environmental performance of Greening-Up strategy: (i) product redesign cost F; and (ii) responsiveness of Traditionals segment customers to the environmental attributes of a product (1 − θ ). Apparently, the former adversely affects the environmental performance of Greening-Up strategy because higher costs deter firms from making incremental changes that would carry product design toward no-impact outcomes. The latter shows that if incremental changes force Traditionals segment customers to compromise on traditional product attributes (e.g., convenience, price and performance), then Greening-Up strategy is unlikely to outperform Greening-Out strategy on environmental sustainability basis—even though a significant portion of customers in the market fall within Fence-Sitters and Greens segments (i.e., r F + r G = 0.88). Interestingly, this outcome can be observed even when customers’ marginal valuations on environmental attributes are less than those on brown attributes (i.e., ve /vb < 1). Ultimately, we look at whether there is a strict dichotomy between having a better economic performance and achieving a higher environmental performance. First, it is observed from Table 3, Table 4 and Proposition 8 that when ve /vb ≤ θ , the firm is never better off creating both bottom-line and environmental benefits. That is, as shown in Proposition 8, in regions A, B and C, Greening-Out strategy always outgreens the Greening-Up strategy, but these two green product strategies are always outperformed economically by Greening-Off strategy. This holds true regardless of how environmental and brown attributes of a product are perceived by Traditionals and Greens segment customers, respectively. However, this does not mean that it is supposed to be an either-or proposition. Notice that as long as ve /vb > θ , the firm can achieve both goals simultaneously, even if it means leaving out serving Traditionals segment customers. For example, in region I where θ ∈ [θ A , 1) and ve /vb ∈ [1, 1/(1 − θ )], even though all three strategies are candidates for better economic performance, a new green product targeted at only Fence-Sitters and Greens segment customers can allow the firm to capitalize on becoming greener. At this point, it is important for the firm to gain an understanding of how responsive the customers are to the environmental and brown attributes in a product and the value equation of products in the mind of customers. In other words, rather than focusing myopically on the strategy specific cost parameters F and N , understanding where a firm really stands on the two dimensions of ve /vb and θ is critical to devise profitable products that are better for the environment. The bad implementations of green product strategies (e.g., replacing the existing brown product with a new green product when ve /vb < θ ) can adversely affect the bottom-line.

6 Concluding remarks Consumers are choosing to buy green products and supporting companies perceived to be green. In response, many consumer goods companies, eyeing these green consumers, are moving beyond environment, filling up the pipeline of greener products targeting the green marketplace. These firms are looking for new ways to leverage green thinking into their

123

Ann Oper Res

product designs. At Clorox, for example, CEO Don Knauss has identified sustainability as one of three core consumer trends with which he wants to align Clorox products. The fact is, most companies encounter a fork in this road to sustainability. Should they offer greener products that represent small enhancements or “tweaks” to their existing brown products by replacing virgin materials with recycled content, designing packaging to be refilled, and so forth? Or should they offer a new green product that is built from scratch and that is unabashedly green? It is not easy to answer. Many green products on the market nowadays represent small tweaks to their existing brown counterparts. For example, washing machines save water and energy by tumbling clothes on a horizontal axis as opposed to a vertical axis. Although these are needed technical achievements as well, our findings underline that the reductions in environmental impact greened-up products represent may not be enough to address consumers’ environmental concerns and so greening up an existing brown product may not be a good option for companies looking to cash in on the potential for green-oriented sales. Rather, it might be innovative green products with radical new designs that would offer companies double bottom-line benefits—one bottom line for profits, the other one reflecting their contribution to environmental cleanup. With this regard, it is important to note that creating bottom-line benefits and reducing environmental footprint is not necessarily an either-or proposition. Both goals can be accomplished simultaneously, but it is important for a firm to gain an understanding of how responsive the customers are to the environmental and brown attributes in a product and the value equation of products in the mind of customers. Another key takeaway is that green consumer segments are complex and selecting the right consumer to target is critical to the success of each green product strategy. Knowing that there are many different types of green consumers, it is essential for company managers to keep in mind that there are many different kinds of environmental issues of concern ranging from hazardous waste to energy efficiency—that concern even the non-environmentalist consumers. There are also opportunities to extend our research in several directions. To start with, we assume no repeat purchases. However, Deloitte recently examined 6,498 shopper experiences in 11 major U.S. retailers and found that green shoppers are a valuable consumer target because they represent a consumer segment who buy more products on each store visit, who visit the store more regularly, and who demonstrate more brand and retailer loyalty in their purchasing behavior.11 They are active consumers who buy more and shop more often as opposed to the average shopper, and they are generally not bargain hunters. More importantly, once a green product has captured these shoppers’ attention, it tends to create brand stickiness by retaining their loyalty through repurchase. That’s why addressing repeat purchase behavior of green consumers would be an interesting extension. Second, we characterize the environmental profile of a product by a single environmental attribute. However, a single green metric might not be amenable to evaluate trade-offs within multiple dimensions of “green” (e.g., energy efficiency, toxicity and recyclability). For instance, CPC—the makers of Mueller’s pasta—found that converting to recycled carton material would actually add about 20 % to the width of their package material and this would partially offset savings to the environment, considering the extra energy needed to ship new boxes. Using a single green attribute can also overlook the fact that consumers’ might have distinct valuations for environmental attributes. Hence, an analysis of a product design decision where a product can be characterized on multiple “green” dimensions would be another potential extension of our study. 11 https://www.deloitte.com/assets/Dcom-Lebanon/Local%20Assets/Documents/Consumer%20Business/ DeloitteGreenShopperStudy_2009.pdf, accessed on March 17, 2013.

123

Ann Oper Res

Another important assumption in our model is that the firm commits in advance to the product and pricing strategy. This implies that before any sales take place, the firm decides on (and commits to) the qualities, the selling price and the type of product he would offer. Moreover, customers believe in these commitments and act accordingly. The firm thus solves the two-period maximization problem only once-at the start of the first period. Here it should be noted that we adopt a two-period setting in our model to factor in changing market segment compositions and market sizes rather than a product and/or pricing strategy. In addition to commitment assumption, we also assume that the firm does not price discriminate between market segments to be in line with prior literature on vertical differentiation. Based on this, one potential extension might be to allow for third-degree price discrimination. Second, firms in some industries (e.g., industries characterized by large production uncertainties) may not be able to credibly commit in advance to a product and pricing strategy, meaning that they can only position and price their products one period at a time (as and when they are offered). Then, it might be interesting to extend our model by considering a firm incapable of commitment, meaning that the firm should choose different green product strategies (and so offer different products) in different periods. Note that in this case the firm must take customer behavior into account when choosing his strategy in each period, because potential customers would rationally look ahead to what the firm might do in period 2 before deciding when to buy.

Appendices Appendix 1: Proof of Proposition 1 Under the Greening-Off strategy, the firm is indifferent between the two pricing options if ∗ ∗ ∗ ∗ and only if Π1i given in (7) is equal to Π1ii given in (10). Solving Π1i = Π1ii for θ shows    2 M2 (r G +δ) , denoted by θ A . that this equality holds as long as θ is equal to 1 − e1 M1er1GM+e 1 +e2 M2  

The proof follows. Appendix 2: Proof of Proposition 2

Under the Greening-Out strategy, the firm is indifferent between the two pricing options if ∗ ∗ ∗ ∗ and only if Π2i given in (14) is equal to Π2ii given in (17). Solving Π2i = Π2ii for θ shows    2 M2 (r T −δ) , denoted by that this equality holds as long as θ is equal to 1 − 1 − e1 M1er1TM+e 1 +e2 M2  

θ B . The proof follows. Appendix 3: Proof of Proposition 3

Note that the solution to the problem stated above is feasible (i.e., both qb3 and qe3 are non∗ negative), if and only if 1 ≤ θvveb ≤ 1 + cFxˆ . Since x1 > x2 , we know that: (a) Π3ii (xˆ = ∗

u

x1 ) > Π3ii (xˆ = x2 ); and (b) cuFx1 < cuFx2 . Therefore, when 1 ≤ θvveb ≤ 1 + cuFx1 , it is optimal to cover all market segments, while the only possibility for the firm is to cover the Fence-Sitters and Greens market segments when 1 + cuFx1 < θvveb ≤ 1 + cuFx2 .  

123

Ann Oper Res

Appendix 4: Proof of Proposition 4 Note that the solution to the problem stated above is feasible (i.e., both qb3 and qe3 are non∗ negative), if and only if 1 ≤ θvveb ≤ 1 + cFyˆ . Since x1 > x3 , we know that: (a) Π3iii (xˆ = u

∗

x1 ) > Π3iii (xˆ = x3 ); and (b) cuFx1 < cuFx3 . Therefore, when 1 ≤ θvveb ≤ 1 + cuFx1 , it is optimal to cover all market segments, while the only possibility for the firm is to cover the Traditionals and Fence-Sitters market segments when 1 + cuFx1 < θvveb ≤ 1 + cuFx3 .   Appendix 5: Proof of Proposition 5

When θ ≤ ve /vb < 1, the second pricing strategy is the only feasible (i.e., both qb3 and qe3 are non-negative) and so the optimal option. When 1 ≤ ve /vb < 1/ (1 − θ ), the first and second pricing strategies become feasible. In this case, the first pricing option dominates the second ∗ ∗ one only if Π3i > Π3ii . On the basis of Proposition 3, comparison of the expressions provided in Eqs. (22) and (26) yields the necessary condition for the above inequality to hold. Finally, when ve /vb ≥ 1/ (1 − θ ), all three pricing options are feasible. In this case, ∗ ∗ ∗ comparing Π3i separately with Π3ii and Π3iii shows that it is always dominated by these ∗ two pricing options. Then, in order to determine under what conditions Π3ii is greater than ∗ iii Π3 , we compare the respective expressions provided in Eqs. (26) and (30), and on the basis of Proposition 4, we obtain the necessary condition provided.   Appendix 6: Proof of Proposition 6 When θ = 0, the total number of pricing options under three strategies reduces to four possibilities. Under the Greening-Off strategy, the only option is to set p = vb qb in which case only Traditionals and Fence-Sitters customers are completely served. Under the GreeningOut strategy, the only pricing option is to set p = ve qe , allowing the firm to cover all segments completely. Under the Greening-Up strategy, there are two options available: p = vb qb +ve qe and p = ve qe . In the former case, only Traditionals and Fence-Sitters customers are served entirely, whereas all segments are completely covered in the latter case. Based on these options, we re-solve the firm’s maximization problem, and obtain the optimal profits for each option by using the parameter settings given in Sect. 5. π1i = (0.27)vb2 π2ii

=

(0.26)ve2

(33) −N

(0.29) (ve − vb )2 π3i = (0.18)vb2 + F (4.28)ve2 ii π3 = 12.42 + 4F

(34) (35) (36)

Mutual comparison of these expressions leads to the respective dominance relationships provided in Proposition 6.   Appendix 7: Proof of Proposition 7 When θ = 1, the total number of pricing options under three strategies reduces to four possibilities. Under the Greening-Off strategy, the only option is to set p = vb qb and completely cover all segments. Under the Greening-Out strategy, the only pricing option is to set p = ve qe , allowing the firm to serve only Fence-Sitters and Greens customers completely.

123

Ann Oper Res

Under the Greening-Up strategy, there are two options available: p = vb qb + ve qe and p = vb qb . In the former case, only Fence-Sitters and Greens customers are served entirely, whereas all segments are completely covered in the latter case. Based on these options, we re-solve the firm’s maximization problem, and obtain the optimal profits for each option by using the parameter settings given in Sect. 5. π1ii = (0.52)vb2 π2i

=

(0.23)ve2

−N

(0.87) (ve − vb )2 π3i = (0.31)vb2 + F π3iii = (0.35)vb2

(37) (38) (39) (40)

Mutual comparison of these expressions leads to the respective dominance relationships provided in Proposition 7.   Appendix 8: Proof of Proposition 8 Based on the environmental qualities provided in Table 4, the aggregate environmental qualities supplied under two pricing options of the Greening-Out strategy are equal if and only if θ = 0.10. Then, when θ ≤ 0.10, it is better for the environment to target the green product at all market segments. Otherwise, targeting the product at Fence-Sitters and Greens segment customers yields higher overall environmental quality. To sum up,  (0.52)(1 − θ )ve , if θ ≤ 0.10, TEQ2 = (41) (0.47)ve , else. Regarding the Greening-Up strategy, as long as ve /vb > θ , setting the price of the greenedup product according to Greens segment customers’ product valuation (and targeting all market segments) always results in more aggregate environmental quality than other two pricing options, in which case TEQ3 = (2.15) (ve − θ vb ) /F.

(42)

Combining the conditions and expressions given above yields the proposition and the proof follows.  

References Anderson, E. T., & Dana, J. D. (2009). When is price discrimination profitable? Management Science, 55, 980–989. Atasu, A., Sarvary, M., & Wassenhove, L. N. V. (2008). Remanufacturing as a marketing strategy. Management Science, 54, 1731–1746. Chen, C. (2001). Design for the environment: A quality-based model for green product development. Management Science, 47, 250–263. Choudhary, V., Ghose, A., Mukhopadhyay, T., & Uday, R. (2005). Personalized pricing and quality differentiation. Management Science, 51, 1120–1130. Debo, L. G., Toktay, L. B., & Wassenhove, L. N. V. (2005). Market segmentation and product technology selection for remanufacturable products. Management Science, 51, 1193–1205. Desai, P., Kekre, S., Radhakrishnan, S., & Srinivasan, K. (2001). Product differentiation and commonality in design: Balancing revenue and cost drivers. Management Science, 47, 37–51. Dobson, G., & Kalish, S. (1988). Positioning and pricing a product line. Marketing Science, 7, 107–125. Ferrer, G. (1996). Market segmentation and product line design in remanufacturing. Working Paper 96/66/TM, INSEAD, Fontainebleau, France.

123

Ann Oper Res Ferrer, G., & Swaminathan, J. M. (2010). Managing new and remanufactured products. European Journal of Operational Research, 203, 370–379. Ginsberg, J. M., & Bloom, P. N. (2004). Choosing the right green marketing strategy. MIT Sloan Management Review, 46, 79–84. Heese, H. S., & Swaminathan, J. M. (2006). Product line design with component commonality and costreduction effort. Manufacturing & Service Operations Management, 8(2), 206–219. Hopkins, M. S. (2010). How sustainability fuels design innovation. MIT Sloan Management Review, 52, 75–81. Katz, M. L. (1984). Firm-specific differentiation and competition among multiproduct firms. The Journal of Business, 57, 149–166. Kim, K., & Chhajed, D. (2000). Commonality in product design: Cost saving, valuation change and cannibalization. European Journal of Operational Research, 125, 602–621. Kim, K., & Chhajed, D. (2002). Product design with multiple quality-type attributes. Management Science, 48, 1502–1511. Klenier, A. (1991). What does it mean to be green? Harvard Business Review, 69, 38–47. Krishnan, V. V., & Lecourbe, P. (2010). Designing product lines with higher aggregate environmental quality. Working Paper. Lacourbe, P., Loch, C. H., & Kavadias, S. (2009). Product positioning in a two-dimensional market space. Production and Operations Management, 18, 315–332. Lancaster, K. (1990). The economics of product variety: A survey. Marketing Science, 9, 189–210. Liu, Z., Anderson, T. D., & Cruz, J. M. (2012). Consumer environmental awareness and competition in two-stage supply chains. European Journal of Operational Research, 218, 602–613. Moorthy, K. S. (1984). Market segmentation, self-selection, and product line design. Marketing Science, 3, 288–307. Mussa, M., & Rosen, S. (1978). Monopoly and product quality. Journal of Economic Theory, 18, 301–317. O’Keeffe, M. (1980). Quality and price discrimination. Ph.D. Thesis, Harvard University. Russo, M. V. (2008). Environmental management: Readings and cases. London: Sage. Souza, G. C. (2013). Closed-loop supply chains: A critical review, and future research. Decision Sciences, 44, 7–38. Subramanian, R., & Subramanyam, R. (2012). Key factors in the market for remanufactured products. Manufacturing & Service Operations Management, 14, 315–326. Wu, L. (2010). Managing design quality and conformance quality: Models and insights. Total Quality Management, 21, 383–389. Yenipazarli, A., & Vakharia, A. (2015). Pricing, market coverage and capacity: Can green and brown products co-exist? European Journal of Operational Research, 242(1), 304–315.

123