Learning from Conformance Quality Failures That Triggered Product Recalls: The Role of Direct and Indirect Experience

• DOI: http://doi.org/10.1111/jscm.12143
• Author: David C. Hall,Tracy D. Johnson‐Hall
• Published date: 01 October 2017
• Date: 01 October 2017

LEARNING FROM CONFORMANCE QUALITY FAILURES

THAT TRIGGERED PRODUCT RECALLS: THE ROLE OF

DIRECT AND INDIRECT EXPERIENCE

DAVID C. HALL

Wright State University

TRACY D. JOHNSON-HALL

College of William and Mary

A product recall is an observable external quality failure and a source of

significant potential loss to firms as well as a threat to public safety. While

such failures can be a motivation for organizational learning, little is

known about the extent to which recalls can be a source of learning expe-

rience. Using econometric modeling, we investigate to what extent firms

learn from conformance quality failures using a panel sample of U.S. food

product recalls conducted by 125 publically traded firms between 2004

and 2013. We differentiate between direct (i.e., internal locus of failure)

and indirect (i.e., supplier locus of failure and industry recalls) sources of

experience and examine the main and interaction effects on the likelihood

of future recalls. The results indicate that experience derived from internal

and supplier failures reduces the potential for future recalls, while indus-

try recalls have no significant main effect. We also find evidence of substi-

tution effects between internal and supplier failures, suggesting that there

are diminishing learning effects at higher levels of both of these types of

failures. This finding suggests that the variety and volume of information

as well as uncertainty regarding outcomes can overwhelm firm informa-

tion processing capabilities at higher levels of these failure types and that

additional resources should be allocated to encourage adequate learning.

Finally, internal failures and industry recalls act as learning complements,

which may indicate that internal quality failures become more relevant

when industry peers also experience quality failures.

Keywords: organizational learning and knowledge acquisition; quality risk; supply

chain management; panel studies; econometric modeling

INTRODUCTION

Organizational learning from quality failure is essen-

tial to improving quality management systems and

reducing the likelihood of future failures (Anderson,

Rungtusanatham & Schroeder, 1994; Sitkin, Sutcliffe,

and Schroeder 1994; Li & Rajagopalan, 1997; Kumar

& Schmitz, 2011). Product recalls remove affected

products from the marketplace to limit consumer risk

and are evidence of significant external quality fail-

ures. Numerous product recalls across multiple indus-

tries in recent years suggest anecdotally that recalls are

a persistent phenomenon with potentially severe

impacts to both firm performance and public safety

(APICS 2016; Cowley, 2016). For example, in the set-

ting for this study, the U.S. food industry, the number

of recall events and affected products has been

increasing over the past decade (United States Food

and Drug Administration, 2015). Moreover, 81 per-

cent of food manufacturers surveyed responded that

the financial risk from a product recall was “signifi-

cant” to “catastrophic” (Grocery Manufacturers of

America 2011, p. 2). In some cases, recalls of defective

food ingredients cascade through a supply chain

(Johnson-Hall, 2017; Ketchen, Wowak & Craighead,

2014), affecting many firms and products downstream

from the original supplier. In 2009, for example,

October 2017 13

Journal of Supply Chain Management

2017, 53(4), 13–36

©2017 Wiley Periodicals, Inc

peanut ingredients from the Peanut Corporation of

America affected approximately 3,900 products at over

200 firms and are confirmed to have caused 9 deaths

and over 700 illnesses (Centers for Disease Control

and Prevention, 2009). Therefore, whether the source

of the defect is a supplier or internal operations, firms

appear to have substantial incentives to learn from

prior recall events in order to reduce the risk of future

recalls.

Extant research on product recalls and organiza-

tional learning suggests that when firms have prior

recall experience, the likelihood of future recalls may

be reduced (Haunschild & Rhee, 2004; Kalaignanam,

Kushwaha & Eilert, 2013; Thirumalai & Sinha, 2011)

or increased (Haunschild & Rhee, 2004; Steven, Dong

& Corsi, 2014). These mixed findings are perplexing

because it could generally be expected that as firms

gain experience with recalls, the incidence of future

recalls by the same firm, hereafter referred to as the re-

calling firm, would be reduced. To shed light on these

mixed findings, we focus on what Argote and Miron-

Spektor (2011) describe as “the most fundamental

dimension of experience” (p. 1126), that is, whether

the recall experience is acquired directly by the recall-

ing firm, or indirectly through the experience of other

organizations.

Accordingly, this study makes multiple contributions

to theory and practice by increasing the granularity

with which the effects of potential learning experi-

ences are evaluated. First, we differentiate between

direct and indirect sources of learning experience and

their effects on learning from quality failures (Argote

& Miron-Spektor, 2011). Specifically, we distinguish

between experience gained from recalls where the

source of the quality problem is internal to the recall-

ing firm, and recalls where the source of the problem

is a supplier to the recalling firm. Direct experience

from internal failures is characterized as an internal

locus of failure, while indirect experience from supplier

failures is referred to as a supplier locus of failure (John-

son-Hall, 2017). We therefore contribute to supply

chain management literature and theory by placing

different experience types in the context of supply

chain relationships and testing their effects on learn-

ing outcomes. Essentially, by understanding the types

of experience that translate into learning subsequent

to quality failure, we contribute to the supply chain

literature that explains the trade-offs that firms make

when they choose between internalizing manufactur-

ing processes versus external sourcing (Cousins, 2005;

Handley, 2012; Leonard-Barton, 1992). Second, we

identify industry recalls as an additional source of

indirect experience. Because managers may perceive

recalls as more relevant when they occur within the

same industry sector, such events may serve as a

source of vicarious learning or knowledge transfer.

Third, we consider the potential for the identified

types of experience to act as substitutes or comple-

ments (Argote & Miron-Spektor, 2011). If, in fact, we

find evidence of different sources of experience acting

as complements or substitutes, we may be able to

clarify the mixed findings of prior studies regarding

the effect of learning experience on future recalls (see

Haunschild & Rhee, 2004; Kalaignanam et al., 2013;

Steven et al., 2014; Thirumalai & Sinha, 2011).

Finally, by studying food product recalls, we exam-

ine the effects of learning experience in a context in

which all of the experiences are related to confor-

mance quality failures, rather than a mix of confor-

mance and design failures. Unlike automotive,

medical device, or consumer product recalls, for exam-

ple, which can be caused by both product design and

conformance quality issues (Bapuji & Beamish, 2008;

Hora, Bapuji & Roth, 2011), food industry recalls are

exclusively the result of conformance quality failures.

The distinction between learning from conformance

quality failures and other types of quality failures is

important, in part, because of the different types of

organizational processes and functional areas involved

in improving design versus conformance quality (Gar-

vin, 1987; Juran, Godfrey, Hoogstoel & Schilling,

1999).

To achieve these research objectives, we investigate

how firm learning outcomes vary based on the differ-

ent sources of recall experience in the context of Food

and Drug Administration (FDA)- and U.S. Depart-

ment of Agriculture (USDA)-regulated food product

recalls occurring between 2004 and 2013. Building on

the broader organizational learning literature in addi-

tion to the product recall literature, we propose the

following research questions: Do organizational learning

outcomes from product recalls vary with direct versus indi-

rect experience? If so, do direct and indirect experience

interact as complements or substitutes to enhance or dimin-

ish organization learning? We address these questions

by evaluating a negative binomial regression model to

predict recalls on a firm-year basis, controlling for

firm and recall attributes (Haunschild & Rhee, 2004;

Kalaignanam et al., 2013; Steven et al., 2014; Thiru-

malai & Sinha, 2011).

Figure 1 summarizes the conceptual model, show-

ing our proposed framework of relationships between

learning experience types and future external confor-

mance quality failures. The remainder of this work is

organized as follows: First, we review the broader

organizational learning and supply chain product

recall literature; we then develop our hypotheses. We

next describe the data sources, sample, methodology,

and then our results. We conclude with a discussion

of the results including implications for theory and

practice as well as potential future research

directions.

Volume 53, Number 4

Journal of Supply Chain Management

14