Optimal pricing for new and remanufactured products

• DOI: http://doi.org/10.1016/j.jom.2015.03.007
• Date: 01 May 2015
• Published date: 01 May 2015

Journal of Operations Management 36 (2015) 130–146

Contents lists available at ScienceDirect

Journal of Operations Management

journal homepage: www.elsevier.com/locate/jom

Optimal pricing for new and remanufactured products

James D. Abbeya,∗, Joseph D. Blackburn b,1, V. Daniel R. Guide Jr.c,2

aDepartment of Information & Operations Management, Mays Business School, Texas A&M University, 4217 TAMU, College Station, TX 77843-4217,

United States

bJames A. Speyer Professor of Production Management Emeritus, Owen Graduate School of Management, Vanderbilt University, 401, 21st Avenue South,

Nashville, TN 37203, United States

cSmeal Chaired Professor of Supply Chain Management, Smeal College of Business, The Pennsylvania State University, 413 Business Building,

University Park, PA 16802, United States

article info

Article history:

Received 17 June 2014

Received in revised form 9 February 2015

Accepted 30 March 2015

Available online 6 April 2015

Accepted by Thomas Younghoon Choi

Keywords:

Closed-loop supply chains

Remanufacturing

Consumer products

Product pricing

abstract

This work investigates the optimal pricing of new and remanufactured products using a model of con-

sumer preferences based on extensive experimentation. The experimental investigation reveals two

distinct segments of consumers. One segment is relatively indifferent between new and remanufactured

products and displays high sensitivity to price discounts. The second segment shows strong preferen-

ces for new products—with an accompanying aversion to remanufactured products—and realtively low

sensitivity to price discounts. The pricing analysis examines several scenarios involving a new product

manufacturer, ranging from a simple monopolist scenario to a more complex scenario involving compe-

tition with third-party remanufacturers. In contrast to the usual finding that new product prices should

decrease when competitive remanufactured products enter the market, the introduction of market seg-

ments reveals a robust finding across all scenarios: when remanufactured products enter the market, the

optimal price of the new product should increase. Through appropriate pricing of new products, the OEM

can mitigate the effects of cannibalization and increase profitability.

© 2015 Elsevier B.V. All rights reserved.

1. Introduction and motivation

The market for remanufactured products is large and grow-

ing. Recent estimates for remanufactured product sales exceed

$100 billion per year with consumer markets representing approx-

imately $10 billion worth of sales per year (Giuntini, 2012; Hauser

and Lund, 2003). Although business-to-business sales still dwarf

consumer sales of remanufactured products, consumer markets for

remanufactured products are growing rapidly, accelerated by the

widespread popularity of Internet sales and on-line auctions. The

increased demand and availability of remanufactured products in

competition with new products pose vexing challenges for the OEM

who manufactures new products.

Faced with competition from remanufactured products, OEMs

struggle to develop a coherent remanufacturing strategy: whether

∗Corresponding author. Tel.: +1 979 862 6515; fax: +1 979 845 5653.

E-mail addresses: jabbey@mays.tamu.edu (J.D. Abbey),

Joseph.Blackburn@owen.vanderbilt.edu (J.D. Blackburn), dguide@psu.edu

(V.D.R. Guide Jr.).

1Tel.: +1 615 322 0645.

2Tel.: +1 814 865 6103.

to enter the market with a remanufactured product and, if

so, how to price the products within a portfolio of both new

and remanufactured offerings. Interviews with managers also

revealed that cannibalization of new product sales by remanu-

factured products represents a significant concern for OEMs. For

instance, at Hewlett–Packard, executives expressed their beliefs

about cannibalization by stating that, “for every four unit sales of a

remanufactured product, they lost one new product sale” (Rysavy,

2001). Other firms have focused on eliminating returns in order

to block third-party remanufacturing and prevent cannibalization

(Blackburn et al., 2004; Ferguson et al., 2006). Even when returns

are unavoidable, firms such as Bosch Tools, NA have hesitated to

offer a remanufactured version of some products for fear of dilut-

ing both their brand and cannibalizing their new product sales

(Valenta, 2008).

Discussions with product managers at OEMs revealed that

the managers often make pricing and other strategic decisions

about remanufactured products in an environment of fear and

uncertainty—fear of cannibalizing new product sales and uncer-

tainty about consumer preferences and willingness to pay for

remanufactured products. Given these concerns and lack of appro-

priate information, some product managers—those who do choose

to offer a remanufactured product—follow ad hoc pricing strategies.

http://dx.doi.org/10.1016/j.jom.2015.03.007

0272-6963/© 2015 Elsevier B.V. All rights reserved.

J.D. Abbey et al. / Journal of Operations Management 36 (2015) 130–146 131

In many cases, those pricing strategies amount to simply pricing

the remanufactured product at some prescribed discount off the

price of a comparable new product through educated guesses or

historical rules of thumb.

To address these concerns, this paper develops models to

help OEMs evaluate the risks of cannibalization, make strate-

gic decisions about entering the remanufacturing market, and to

provide guidance on product pricing—the joint pricing of the new

and the remanufactured products. Though a sizeable literature

on remanufactured consumer products exists, most models use

standard, theoretical assumptions about consumer preferences for

new and remanufactured products that have not been fully vali-

dated (Watson, 2008). Further, Guide and Van Wassenhove (2009)

raised concerns about the validity of these theoretical consumer

behavior models and called for empirical studies of consumer pre-

ferences for both new and remanufactured products. In light of

these concerns, the empirics of Section 3investigate consumer

preferences and reveal distinct departures from the classical theo-

retical model assumptions. Those departures from the assumptions

of the classical theoretical models reveal deficiencies in the exist-

ing literature that limit the existing models’ effectiveness for use

in practice. Based on the empirical findings, the resulting models

of Section 4are quite distinct from previous classical theoretical

models. As a result, this study differentiates itself from previous

research and provides constructive new insights for practitioners

regarding consumer preferences for the products.

Based on these concerns and open questions, this research has

twin objectives. The first objective is to gain empirical understand-

ing regarding the nature of consumer preferences for new and

remanufactured products. The second objective is to use the empir-

ical findings of consumer behavior as a means to inform practice

on remanufacturing strategy. Those strategies provide practition-

ers with insights for improved pricing/production decisions when

remanufactured products are introduced to the market. The empir-

ical analysis, detailed in Section 3, addresses the first objective

through a tailored experiment, using a general population of U.S.

consumers, to facilitate development of a new analytical model of

consumer preferences for new and remanufactured products. This

model differs in significant ways from the consumer behavior mod-

els that have been the de facto standard in the remanufacturing

literature. In particular, through use of a highly generalizable mixed

between and within-subject mixed model with subsequent regres-

sion analyses, the experimental work identifies distinct segments

of the consumer population that have markedly different prefer-

ences for remanufactured products. Providing insights for practice

through improved strategy— the second objective—requires use

of the consumer behavior model as input into an analytical pri-

cing model. The analysis of this empirically informed analytical

model constitutes the second stage of the research. This second

stage, covered in Section 4, develops market entry and pricing pol-

icy decisions under different scenarios that capture the variety of

market conditions an OEM would face in competition with reman-

ufactured products. The analytical results provide new insights

into the appropriate discount levels for remanufactured products

and yield counter-intuitive results for pricing new products. The

results show that optimal pricing for new products often embodies

price-increasing competition: competition from the remanufac-

tured product causes the optimal price of the new product to

increase, rather than decrease.

2. Related literature

In addition to the research cited in Section 1, this manuscript

draws on an extensive body of literature in consumer behavior,

market segmentation, and remanufacturing strategy. In doing so,

this manuscript addresses two fundamental elements of reman-

ufacturing strategy. The first fundamental element addresses

optimal pricing for the two products when a new product and a

remanufactured product compete in the same market. The sec-

ond fundamental element studied identifies the optimal level of

production if the firm decides to offer both new and remanu-

factured products simultaneously. Previous research addressing

these questions has followed a methodology that has its roots in

classical market segmentation economics. Specifically, the mar-

ket segmentation studies analyze the price and positioning of a

portfolio of products by first positing a theoretical model for the dis-

tribution of consumer preferences, then using that assumed model

to determine the optimal prices and positions under competitive

equilibrium (Hotelling, 1929; Mussa and Rosen, 1978; Moorthy,

1984; Vandenbosch and Weinberg, 1995).

Although the market segmentation literature provided the

research template used in theoretical studies of remanufactur-

ing strategy, there are important differences between the two

problems. Remanufacturing decisions are structurally simpler than

those of the classical market segmentation problem because, with

remanufacturing, the product positioning decisions are fixed a pri-

ori. In other words, there are existing new products with defined

positions and a remanufactured product—functionally equivalent

to the new product that may also enter the market. In addition,

the realities of a closed-loop supply chain impose constraints on

the production quantities of remanufactured products. The amount

of remanufacturable cores is limited to a fraction of prior produc-

tion based on a firm’s reverse supply chain and product acquisition

management (Guide and Van Wassenhove, 2001).

The seminal paper by Majumder and Groenevelt (2001) initiated

a stream of research on remanufacturing strategy. The Majumder

and Groenevelt paper, and most of the subsequent papers on

remanufacturing strategy, adhere to traditions of market segmen-

tation research by beginning with an assumed theoretical model of

consumer behavior. These theoretical models assume that all con-

sumers would buy a remanufactured product at a sufficiently large

discount on some distribution—usually a uniform distribution—of

willingness-to-pay preferences across an interval of prices. With

few exceptions, these models generally employ a linear relation-

ship between the discount and the consumer’s preference for

remanufactured products (Atasu et al., 2008; Debo et al., 2005;

Jin et al., 2007; Vorasayan and Ryan, 2006). In all of these studies,

the consumer behavior model finds the competitive equilibrium

to determine the price and production quantities for the new

and remanufactured products. In related work, Aydinliyim and

Pangburn (2012) consider a problem that is similar in structure to

the remanufacturing strategy problem. In their model, consumers

choose between two products that differ only in their packaging

with reduced-packaging products offered at a discount. Unlike the

remanufacturing strategy models, the theoretical consumer behav-

ior model used in their work allows for the possibility that the

willingness-to-pay functions may intersect. In other words, some

fraction of consumers (e.g., green consumers) might be willing to

pay a premium for the reduced-packing option.

In contrast to the models based on the common set of theo-

retical assumptions, motivation for this manuscript stemmed from

the growing body of empirical evidence that the theoretical mod-

els of consumer behavior used in prior studies may be deficient

and fail to capture accurately the complexity of consumer pre-

ferences. A recent field experiment using eBay auctions of both

new and remanufactured products found that consumers tended

to fall into one of two distinct segments (Guide and Li, 2010). One

segment only bid on new products and stated in follow-up sur-

veys that they would not consider purchasing a remanufactured

version. The second segment contained deal-seeking, functional

consumers who displayed relative indifference between new and