The current issue and full text archive of this journal is available on Emerald Insight at:

www.emeraldinsight.com/0144-3577.htm

Supply chain
Supply chain resilience: the whole resilience
is not the sum of the parts
Marcelo Martins de Sá and Priscila Laczynski de Souza Miguel
Department of Operations Management, Fundacao Getulio Vargas,
Sao Paulo, Brazil
Renata Peregrino de Brito Received 4 September 2017
IAG Business School, Revised 5 March 2018
2 July 2018
Pontificia Universidade Catolica do Rio de Janeiro, Rio de Janeiro, Brazil, and 14 September 2018
Susana Carla Farias Pereira 11 December 2018
11 January 2019
Department of Operations Management, Fundação Getulio Vargas, 12 February 2019
Accepted 19 February 2019
São Paulo, Brazil

Abstract
Purpose – The purpose of this paper is to investigate how resilience at different nodes in the supply chain
influences overall supply chain resilience (SCRES) during an extreme weather event.
Design/methodology/approach – Based on 41 in-depth interviews, this qualitative study examines two
Brazilian agri-food supply chains (AFSC). The interviews explored the impacts, preparedness, response
and adaptation strategies adopted by farmers, processors and manufacturers during Brazil’s extreme
drought of 2014–2015.
Findings – SCRES does not depend on all organizations in the supply chain but rather on the company able
to reconfigure the resources to control for the disruption. In a supply chain with low interdependence among
players, individual firm resilience elements might be preferable to interorganizational ones.
Research limitations/implications – This study is based on the context of AFSCs with low
interdependence among players and during the experience of a climatic event. The results might not be
generalizable to other sectors and phenomena.
Practical implications – Firms must evaluate their positions in supply chains and their interfirm
relationships to determine which resilience strategy to invest in and rely on. Moreover, to leverage resilience
at the supply chain level, firms must intensify information sharing and improve proactive resilience strategies
upstream as well as downstream in the supply chain.
Originality/value – This study presents a broader perspective of resilience by comparing resilience
elements at both the node and supply chain levels and by discussing their interactions and trade-offs.
Keywords Resilience, Case study, Agri-food supply chain
Paper type Research paper

1. Introduction
The increasing frequency and impact of unexpected adverse events have led researchers
and practitioners to shift from a traditional risk management approach to the resilience
approach ( Jüttner and Maklan, 2011; Pettit et al., 2013). Resilience enables systems to cope
with the unexpected (van der Vegt et al., 2015) and to ensure continuity of operations and
delivery to final customers (Christopher and Peck, 2004; Ponomarov and Holcomb, 2009;
Stone and Rahimifard, 2018). Although resilience seems to engage all organizations in a
system, rarely has it been investigated how each part contributes to the overall process,
such as in the context of supply chains.
Since organizations are interconnected, their interdependencies can magnify the
consequences of local events and cause disruptions across the whole supply chain
( Jüttner et al., 2003; van der Vegt et al., 2015; Stone and Rahimifard, 2018). Building
resilience is strategic and helps organizations anticipate disruptions and adapt to new International Journal of Operations
& Production Management
post-event states (Ponomarov and Holcomb, 2009; Ali et al., 2017). However, resilience © Emerald Publishing Limited
0144-3577
strategies implemented at individual nodes can be detrimental to upstream or downstream DOI 10.1108/IJOPM-09-2017-0510
IJOPM stages and may not contribute to the creation of a stronger supply chain (Tukamuhabwa
et al., 2015; Stone and Rahimifard, 2018). Despite that, the literature on supply chain
resilience (SCRES) has focused on the organizational level ( Jüttner, 2005), particularly on
focal companies, which are usually in a position to structurally influence the supply chain
(Christopher and Peck, 2004; Wieland and Wallenburg, 2013; Brandon-Jones et al., 2014;
Tukamuhabwa et al., 2017). Thus, while investigating the effects from the perspective of
focal firms, the study of SCRES might overlook consequences for the overall supply chain.
Another gap relates to the development of resilience elements at the different phases of
preparedness, response, recovery and adaptation (Hohenstein et al., 2015; Tukamuhabwa
et al., 2015; Kamalahmadi and Parast, 2016; Ali et al., 2017). Resilience elements are
management practices that support SCRES capabilities, such as flexibility, redundancy,
collaboration, visibility and velocity (Ponis and Koronis, 2012; Hohenstein et al., 2015;
Tukamuhabwa et al., 2015). Each phase requires the development of different elements that
can be implemented by firms individually or at the supply chain level. To date, there is no
empirical evidence that all resilience elements are relevant in all different supply chains
(Stone and Rahimifard, 2018). Additionally, little is known about how interfirm relationships
influence the development of resilience elements at each phase of a disruption.
This study addresses these gaps by investigating SCRES from the perspective of
different supply chain nodes, instead of addressing the singular point of view of a focal firm.
Hence, this study poses the following research question:
RQ1. How does resilience, built at different nodes of a supply chain, influence SCRES?
For that purpose, we conducted a qualitative, inductive study regarding the process of
resilience building in the sugarcane and orange supply chains during an extreme drought in
the Southeast region of Brazil. Climatic events are associated with high levels of uncertainty.
The agri-food supply chains (AFSC) is particularly sensitive to these uncertainties, which
requires building resilience. Thus, based on 41 in-depth interviews with farmers, processors,
manufacturers and relevant stakeholders, our research analyzed the resilience at each node
of the supply chain as well as the SCRES and compared the resilience elements adopted by
each node at each resilience phase.
The contribution of our study is twofold. First, this study brings a broader system-wide
perspective of SCRES that addresses the implications for all firms in upstream and
downstream nodes, rather than a focal firm perspective. By investigating three different
nodes in two AFSC, the research reveals the interaction between firm and SCRES and the
contribution of each node to the overall process, providing a more holistic understanding of
SCRES. Our findings highlight the role of different nodes for SCRES, even when upstream
organizations are not resilient. SCRES does not necessarily mean that every organization in
the supply chain is resilient; rather, it is more dependent on one company’s ability to
reconfigure resources and players to control for the disruption.
Second, driven by the AFSC context, this study empirically explores the influence of
interfirm relationships on the development of certain resilience elements at each resilience
phase. SCRES literature has focused primarily on the most cited resilience elements, without
adapting it to different contexts (Stone and Rahimifard, 2018). By comparing resilience
elements at both the node and supply chain level as well as at each phase of a disruption, our
results provide evidence about the importance of considering supply chain context
specificities when making investment decisions for building resilience.
This paper is organized as follows. The first section presents a literature review on SCRES.
Next, data collection and data analysis are described in the Methods section. The Findings
section describes the concept that emerged from the data, and in the Discussion section,
propositions are formulated. Finally, in the Conclusions section, academic and managerial
implications, limitations, and possibilities for future research are presented.
2. Literature review Supply chain
2.1 Firm and supply chain resilience resilience
Resilience is widely defined as the ability of organizations and supply chains to plan for,
respond to, and recover from disruptions in a timely and cost-effective manner; it is the
ability to take actions that should return them to an original or perhaps better state than
before the disruption (Ponomarov and Holcomb, 2009; Wieland and Wallenburg, 2013;
Tukamuhabwa et al., 2015). This definition covers the moments before, during and after the
disruption (Sheffi and Rice, 2005; Ali et al., 2017) and encompasses different phases of
resilience: preparedness, response, recovery and growth or adaptation (Hohenstein et al.,
2015; Ali et al., 2017; Stone and Rahimifard, 2018).
The concept of resilience addresses different units of analysis: the firm and the supply
chain (Kamalahmadi and Parast, 2016; Ali et al., 2017). Firm resilience refers to an
organization’s capacity to anticipate, prepare for, quickly respond to, and then recover from
a crisis by learning from the experience and adapting to the new scenario (Hohenstein et al.,
2015; Ali et al., 2017; Stone and Rahimifard, 2018). Resilient firms are less vulnerable to
disturbances and better able to manage internal resources, such as routines and systems, to
cope with unexpected disturbances (Ponomarov and Holcomb, 2009; Ambulkar et al., 2015).
Firms depend on individuals, processes and organizational culture to build resilience
(Bhamra et al., 2011; Kamalahmadi and Parast, 2016).
SCRES, in turn, refers to how supply chain stakeholders contain and control a
disturbance from spreading to other organizations within the system. In particular, SCRES
focuses on maintaining the core function of supplying goods to end consumers by
anticipating disruptions and developing strategies to decrease a disturbance’s impact, thus
providing immediate response and recovery (Kamalahmadi and Parast, 2016; Ali et al.,
2017). Despite the acknowledged interdependence between the firm and SCRES, the SCRES
literature has not thoroughly investigated the interaction between these two levels.
Supply chains are dynamic systems that connect different organizations, which means
SCRES is impacted by these interfirm relationships, the compatibility amongst firms, the
number of players and the length of the supply chain (Tukamuhabwa et al., 2015;
Kamalahmadi and Parast, 2016; Kaufmann et al., 2018). In this sense, the supply chain
context may influence firm-level decision-making during the crisis, which affects the
SCRES level (Burnard et al., 2018; Stone and Rahimifard, 2018).
For example, transactional and arm’s length relationships do not encourage information
sharing or the development of cooperative and cohesive endeavors among partners, which
are important features of SCRES (Liu et al., 2009; Kaufmann et al., 2018). On the other hand,
interfirm efforts increase supply chain response capacity (Christopher and Peck, 2004;
Revilla and Saenz, 2017). Consequently, it is important to investigate the development of
resilience elements under different contexts of interfirm relationships.

2.2 Resilience elements

Current SCRES research has analyzed what management and operational elements a firm
can adopt to build resilience capabilities (Hohenstein et al., 2015; Ali et al., 2017; Stone and
Rahimifard, 2018). The most cited resilience elements in the literature are flexibility,
redundancy, collaboration, velocity and visibility (Ponis and Koronis, 2012; Hohenstein et al.,
2015; Tukamuhabwa et al., 2015). Flexibility refers to how easy it is for a supply chain to
change based on its range of options (Stevenson and Spring, 2007; Ali et al., 2017). Flexibility
can be achieved by using multiple suppliers’ strategies, flexible supply bases and flexible
processes for operations, transportation and order fulfillment (Kamalahmadi and Parast,
2016; Ali et al., 2017). Redundancy involves maintaining an idle response capacity to access
in case of disruptions. This is done primarily through investments in capital and capacity,
such as safety stocks and backup suppliers or storage (Rice and Caniato, 2003;
IJOPM Kamalahmadi and Parast, 2016). Collaboration refers to the level and degree of shared
information, knowledge and decisions between two or more members of the supply chain
(Christopher and Peck, 2004; Wieland and Wallenburg, 2013; Scholten and Schilder, 2015).
Both vertical and horizontal forms of collaboration are based on trust and synergies (Cao
and Zhang, 2011; Leat and Revoredo-Giha, 2013; Ali et al., 2017). Velocity encompasses the
ability to react rapidly to changes (Christopher and Peck, 2004; Brandon-Jones et al., 2014),
while visibility addresses the extent to which supply chain participants have access to or
share information regarding their operations during the crisis (Barratt and Oke, 2007;
Wieland and Wallenburg, 2013).
Each phase of a disruption requires different elements. Ali et al. (2017) have proposed a
framework that organizes these elements and the supporting managerial practices into
different resilience phases. During the preparedness phase, companies should anticipate the
situation by evaluating and interpreting possible risks, while simultaneously building
redundancy to develop robustness and a security plan. Firms should also increase visibility
by monitoring structures, processes and information at all nodes in the supply chain. During
disruptions, companies must use their flexibility, redundancy and velocity capacities, as
well as use collaboration with other players to respond to an event. After the disruption,
organizations should review and learn from the event and then prepare new contingency
plans for future occurrences.
Resilience elements also differ in scope. Some elements are adopted to manage
disruptions at the organizational level, like firm flexibility and redundancy. Others address
collective efforts to prevent supply chain disruptions, such as collaboration and visibility
(Stone and Rahimifard, 2018). According to Stone and Rahimifard (2018), it is important to
optimize solutions at each resilience phase; however, solutions at each level need to consider
the entire system.
Given the lack of empirical validation, there is poor consensus regarding how these
elements interact to build SCRES. The individual firm’s response to any event depends on
the context in which the organization is embedded, its own resource configurations, as well
as the interactions among organizations and with the natural environment (Tukamuhabwa
et al., 2015; Burnard et al., 2018; Stone and Rahimifard, 2018). Thus, SCRES research should
go beyond simply identifying constructs and begin to explore their interactions. This means
considering how they might complement, reinforce, or conflict with each other
(Tukamuhabwa et al., 2015; Kamalahmadi and Parast, 2016; Stone and Rahimifard, 2018).
In summary, our research aims to fill two gaps identified in the literature:
(1) investigate resilience at the node and the supply chain level to explore the role of
different organizations in the SCRES; and
(2) analyze the influence of interfirm relationships on the resilience elements adopted by
each node at each resilience phase.

3. Methods
This study applies a multiple-case method to investigate how firms at each node of
the AFSC built resilience during a major drought and its overall impact on SCRES.
Given the specificity of the phenomenon investigated in the AFSC, the case study is the
most suitable method (Eisenhardt, 1989; Barratt et al., 2011). Moreover, the inductive
approach is appropriate for building and refining theory regarding specific research
questions and settings such as this one (Eisenhardt, 1989).
The central phenomenon of this study, the extreme drought, took place during the summer
of 2014–2015. One of the most impacted industries was agriculture, particularly the orange
and sugarcane plantations located in the southeast region of Brazil. Given the extensive
impact on agricultural activities, this study investigates the effects of the extreme drought Supply chain
through all nodes of the AFSCs, starting upstream with the farmers. resilience
A generic AFSC includes farmers, processors, manufacturers and distributors of food
(Roth et al., 2008; Stone and Rahimifard, 2018). Unlike a traditional linear supply chain, it can
include thousands of firms, especially upstream in the supply chain (Stone and Rahimifard,
2018). In contrast to most supply chain studies, AFSCs are not based on downstream focal
firms, but rather on upstream firms, as each supply chain has its own characteristics
depending on the product (Maloni and Brown, 2006; Roth et al., 2008).
The AFSC provides a unique context in which to study the interactions of different
resilience elements since it involves organizations which are vulnerable to different, yet
interconnected risks and must constantly adapt to a changing environment (Roth et al.,
2008; Stone and Rahimifard, 2018). Each node in the supply chain introduces different
vulnerabilities that could impact the overall SCRES, such as a product’s shelf life or
specific climatic and environmental risks (Leat and Revoredo-Giha, 2013; Stone and
Rahimifard, 2018). Therefore, we have adopted an embedded, multiple-case design in
which the firms become part of a larger unit, the supply chain node, which is our unit of
analysis (Yin, 1989). Figure 1 illustrates the two supply chains analyzed and their
respective nodes.

3.1 Sugarcane and orange supply chains

Sugarcane’s economic and sociocultural relevance for Brazil goes back to the fourteenth
century. Operations have evolved to produce not only quality sugar but also biofuels, bioenergy
and bioplastics, among others. Today, there are approximately 370 sugarcane mills and more
than 70,000 independent sugarcane farmers (UNICA, 2015). The development of research
centers, industry associations and government agencies has helped to achieve product
Generic AFSC

Node 1: Node 2: Node 3:

Farmers Processors Manufacturers
Sugar Supply Chain

Cachaça_A SugarMill_I
Cachaça_C Bioenergy SugarMill_V
Cachaça_D SugarMill_B SugarMill_Z
SugarFarm SugarMill_C Cachaça_Big
SugarMill_E SugarCoop
F&B_Large
F&B_MNE
F&B_New
Orange Supply Chain

F&B_Small

OrangeProc_A ConCoop
OrangeFarm OrangeProc_B LargeCoop
OrangeProc_C LocalCoop

Figure 1.
Associations and Government Entities
Sugar Bioenergy Association Gov Assistance
Analyzed supply
Orange Association Gov Research chains
IJOPM flexibility in the processing phase. Typically, farmers sell sugarcane to mills (processors)
which sell the sugar either via a cooperative for export or directly to manufacturers.
Relationships between farmers and processors are based on formal contracts. Spot market
purchases are not a typical alternative, and most processors are vertically integrated (producing
up to 80 percent of their own supply).
In 2007, sugarcane farmers and processors in the state of São Paulo signed the
Environmental Protocol and committed to mechanizing the harvesting process, in order to
avoid the practice of burning sugarcane, and to reduce the use of water in processing
(UNICA, 2015). Over the last decade, farmers and processors have adapted their procedures
to the new regulation: in farming, new types of seedlings and new ways of planting
sugarcane were developed; in processing, sugarcane mills implemented closed-circuit
processes to reuse water by treating (regenerating) effluents to reuse either in the mills or on
the plantations (Mosqueira-Salazar et al., 2013).
Brazil is the world’s number one producer of frozen concentrated orange juice, and
most of it is produced in the state of São Paulo (Neves et al., 2013). Orange farmers
sell products to processors (cooperatives) that either resell the products in natura or
produce orange juice. In the orange supply chain, contracts between farmers and
processors are based on quality and volume requirements, making it more difficult for
smaller farmers to compete (Neves et al., 2013). Recently, the orange sector was
consolidated into fewer, larger integrated processors.

3.2 Case study data collection

The selected cases studies consist of 28 organizations. A total of 24 firms are grouped into
three nodes: farmers (5 firms), processors (15 firms, 10 of them vertically integrated) and
manufacturers (4 firms). Additionally, we included sector associations (two organizations)
and government entities (two organizations). To determine the theoretical sampling, we
selected respondents from the first node ( farmers) of the impacted region based on their
ability to provide information regarding the event (Eisenhardt, 1989). The first respondents
in this node were identified with help from industry associations. As the interviews
progressed, they were asked to provide additional names based on their supply chain
relationships so that the impacts could be followed along the supply chain.
The selected cases bring diversity in terms of firm size and structure (vertical integration),
which are relevant aspects in the capacity to build resilience (Sullivan-Taylor and Branicki, 2011).
Table I presents descriptions of the selected firms and their positions in the supply chain as well
as the respondents’ profiles.
The interviews took place during two separate periods after the event – July to
November 2015 and the same timeframe in 2016. The interviews followed a protocol,
involving semi-structured questions (Appendix) regarding broad aspects of the event
impacts, the measures undertaken to adapt operations, the involvement of supply chain
partners, and the plans and preparations for future events. During the initial data collection,
the results were analyzed, and the protocol was adjusted with additional questions to
increase our understanding of the phases of SCRES (Ali et al., 2017). In the second round of
interviews, the initial cases were revisited to complement and clarify the initial responses.
A total of 41 semi-structured interviews were conducted.
Interviewees agreed to have the interviews recorded. The respondents were key
informants responsible for operations at the firms and were involved in the supply chain
relationships. In some cases, the interviews were conducted with more than one respondent,
which allowed us to validate the information. Whenever possible, two researchers
conducted the interviews to increase internal validity (Eisenhardt, 1989; Barratt et al., 2011).
For triangulation purposes, we collected additional data from documents and web sources
(e.g. websites, reports and organization newsletters). Finally, we interviewed industry
Supply Respondents job function Interviews
Supply chain
chain SC node Name Product Size (No. of interviews) length resilience
Sugarcane Farmer Cachaça_A Cachaça Micro Owner (2) 43/32 min
Cachaça_C Cachaça Micro Owner (2) 38/45 min
Cachaça_D Cachaça Micro Owner (1) 25 min
SugarFarm Sugar/ Medium Owner (1) 27 min
Ethanol/
Bioenergy
Farmer/ Bioenergy Ethanol/ Large Managers: Supply Chain (2) 30/18 min
Processor Bioenergy and Legal (2)
Cachaça_Big Cachaça Large Managers: Agriculture (1), 58/82/20
Industrial (1) and min
Sustainability (1)
SugarMill_B Sugar/ Large Operations Director (1) 45 min
Ethanol/
Bioenergy
SugarMill_C Sugar/ Large Operations Director (1) 32 min
Ethanol/
Bioenergy
SugarMill_E Sugar/ Large Agricultural Manager (2) 85/25 min
Ethanol/
Bioenergy
SugarMill_I Sugar/ Large Industrial Manager (1) 35 min
Ethanol/
Bioenergy
SugarMill_V Sugar/ Large Industrial Manager (1) 51 min
Ethanol/
Bioenergy
SugarMill_Z Sugar/ Large Production Manager (1) 37 min
Ethanol/
Bioenergy
Processor SugarCoop Sugar/ Large Sustainability Manager (2) 76/50 min
Ethanol/ and Institutional Advisor (2)
Bioenergy
Orange Farmer OrangeFarm Oranges Small Owner (2) 20/21 min
Farmer/ OrangeProc_A Oranges Large Operations Manager (3) and 25/32/40
Processor Operations Director (3) min
OrangeProc_B Oranges Medium Owner (1) 25 min
Processor OrangeProc_C Oranges Medium Operations Manager (1) 21 min
ConCoop Oranges Small Agricultural Manager (1) 32 min
LargeCoop Oranges Large Agricultural Manager (2) 44/49 min
LocalCoop Oranges Small Operations Director (1) 30 min
Sugarcane Manufacturer F&B_Large Food Large Sustainability Manager (1) 43 min
and F&B_MNE Beverage Large Purchasing Manager (1) 36 min
Orange F&B_New Beverage Large Quality Manager (1) 27 min
F&B_Small Beverage Small Operations Manager (1) 23 min
Associations and government
Sugarcane Association SugarAssoc n/a n/a Water Resources 39 min
Consultant (1)
Orange Association OrangeAssoc n/a n/a Agricultural Manager (1) 32 min
Sugarcane Government GovAssist Assistance n/a Directors: Agricultural (3) 106/72/35
and Engineering (3) min
Orange
Sugarcane Government GovResearch Research n/a Agricultural consultant 30 min
and and Table I.
Orange assistance Cases studied
IJOPM associations and government entities for a better understanding of the AFSCs and to
confirm our findings. This study applied several measures to ensure the rigor of data
collection and analysis, as well as the credibility of our findings (Table II).

3.3 Data analysis

The process of data analysis was iterative throughout the data collection, and the results
were constantly compared to the existing literature to determine the categories that should
be expanded (Eisenhardt and Graebner, 2007). The data analysis followed the inductive
method described by Gioia et al. (2013), and it evolved in three stages, resulting in the data
structure presented in Figure 2.
In the first stage, an open coding process (in vivo coding) of the interviews was adopted
by each researcher, using the informants’ own expressions and terms. During this process,
attention was paid to how firms were impacted by the unexpected event, and how they
responded to it using their own resources or those of the supply chain. The researchers’
coding schemes were shared and discussed by the research team to reach a consensus in
terms of interpretation. The selection of codes was based on their relevance and
predominance during the interviews. For example, “nothing to do” was a typical response to
the event. After comparing the interviews, the codes were refined, and we began grouping
and categorizing similar ones, resulting in the first-order categories (Gioia et al., 2013).
The composition of the first-order categories is demonstrated in the Findings section, with
representative quotes for each category.
In the second stage, we compared the first-order categories to the theoretical concepts
from the literature to consolidate them into the second-order themes (Gioia et al., 2013).
For example, we identified the practices of “multi-sourcing” and “conserving water” in
response to the event as evidence of flexibility. In terms of preparation, the attitudes of
“surprise”, “climate perception” and “lack of information” configured the state of low risk
awareness. “Friction in the supply chain” and “contracts” at different nodes characterized
the power asymmetry of the supply chains context.

Trustworthiness criteria Study method

Credibility (extent to which the results appear to Used the research protocol and definition of questions
be acceptable representations of the data) identified in the literature
Two researchers collected data, and four researchers
analyzed the data
3-page summary of initial interpretations was provided to
the respondents for feedback
Transferability (extent to which the findings Theoretical sampling based on the food supply chain in
from one study in one context will apply to other two distinct industries: sugarcane and orange
contexts) Sample presented firms from different nodes of the supply
chain and different sizes
Dependability (extent is unique to time and place; Respondents recounted experiences covering the
the stability or consistency of explanations) phenomenon studied (water crisis), as well as historic data
from other severe events
More than one informant per firm
Triangulation with associations and government, as well
as with documents
Confirmability (extent to which interpretations All of the interviews and documents were analyzed by
Table II. are the results of the participants and the four co-researchers
Trustworthiness phenomenon, as opposed to researcher biases) Summary of preliminary findings was analyzed by other
of the study and team members who acted as controllers
the findings Sources: Based on Flint et al. (2002), Kaufmann and Denk (2011)
 First-order categories Second-order themes Aggregate dimensions
Supply chain
Productivity decrease resilience
Restricted water supply Direct impact
No significant impact Impact of the event
Increased product cost
Indirect Impact
Lower quality

Node Level
Surprise
Climate perception Low awareness Preparedness

Lack of information

Nothing to do No response

Conserve water Flexibility Response

Multi-sourcing Redundancy
No prevention
No adaptability
Negative cost benefit Adaptation
Willingness to adapt
New investments

No dialogue Low visibility SC preparedness

Supply chain Level

No support No Collaboration
Cooperatives’ support SC response
Collaboration between buyer–supplier Collaboration

Friction in the SC
Contracts Power asymmetry
SC context Figure 2.
Fragmented sector Lack of cohesion Data structure
Sector consolidation

We also compared the codes across nodes during the different phases of the resilience
process to investigate their impact at the supply chain level. This analysis provided a better
understanding of the differences among nodes ( farmers, processors and manufacturers)
as well as the differences between supply chains (sugarcane and orange).
The second-order themes were aggregated into six broader theoretical dimensions that
served as foundations for our discussion (Strauss and Corbin, 1998; Gioia et al., 2013) and are
linked to each phase of resilience and to the supply chain context.
The aggregate dimensions were named: impact of the event, preparedness of the firms,
responses of the firms, adaptation for the future, preparedness of the supply chain,
response of the supply chain and supply chain context. The impact of the event dimension
refers to the direct impact of the event for the different nodes and also reflects
how the direct impact at one node affected all other stages in the chain. The preparedness
of the firms dimension emerged as the degree to which respondents were aware of
the risk of drought and their capacity to anticipate the disruption. Responses of the firms
relate to the strategies chosen and decisions made by firms to manage and adjust their
resources and minimize the negative impact. Adaptation for the future describes the
degree to which different nodes learned from the crisis and whether they were willing
to adapt for future occurrences. The preparedness of the supply chain dimension refers to
collective efforts to exchange information and prepare the supply chain to maintain the
flow of goods, while the response of the supply chain relates to how organizations jointly
planned and coordinated processes to address the drought. Finally, the supply chain
context refers to the structure of the supply chain in terms of the players, the industry
and their relationships.
Relationships between the emergent concepts were then summarized in a model that
illustrates the phenomenon and became the foundation for the propositions presented in the
Discussion section.
IJOPM 4. Findings
This section begins by describing our findings on how the drought impacted each node and
the supply chain. It then examines the node findings by providing an overview of how each
node managed the unexpected event on three aggregate dimensions: preparedness of the
firms, responses of the firms and adaptation for the future. In the supply chain findings
section, results were compared to generate an understanding of the three aggregate
dimensions – preparedness of the supply chain, response of the supply chain and supply
chain context – and the two supply chains are then compared.

4.1 Impact of the event

Analysis of the drought’s impact focused on direct effects (node level) as well as indirect
effects (supply chain level) (Table III). The perceived direct effects were decreased
productivity and restricted water supply. Farmers suffered major losses as in terms of
decreased productivity. Processors suffered from restricted water supplies but recognized
that the major impacts were to farming activities. In contrast, manufacturers did not report
relevant negative impacts on their operations; instead, the water shortage caused an
increase in the sales of soft drinks:
And as a result, we had a large loss in productivity, around 40% […] (SugarFarm).
What we noticed was that we had no impact […] in fact, we were favoured because my sales
increased […] (F&B_MNE).
The drought’s indirect impacts were the result of spillover from other nodes. Decreased
farming productivity resulted in supply shortages for sugarcane and orange supply chains
and increased costs for orange processors. Orange processors also perceived deterioration in
fruit quality.
Overall, the drought resulted in farmers suffering major losses. Processors were less
affected, and manufacturers were not impacted by the drought. In both the sugarcane and
orange supply chains, the upstream impacts did not cause disruptions for end customers,
suggesting that the downstream nodes were resilient. The next sections aim to clarify this
finding by providing a detailed analysis of each node’s resilient phases.

4.2 Node findings

4.2.1 The preparedness of firms. The findings documented three main first-order categories
of low awareness. We have categorized the reactions and attitudes of organizations toward
the unexpected drought as follows: surprise, climate perception and lack of information
(Table IV ). In general, the respondents were not monitoring climatic changes beyond the
very short term. They had no knowledge of eminent climate risks and vulnerabilities, which
resulted in low awareness and did not help them to anticipate the event.
First, despite the climate sensitive aspects of AFSC, farmers were surprised by the
extreme weather event and, therefore, were not prepared for it. Weather is regarded as an
“act of God,” and they did not realize how serious the situation was until they were fully
impacted. Processors were also surprised by the event. While processors recognized climate
as a well-known risk in farming, it was the first time the water restriction impacted the
production capacity of the mills in the sugarcane supply chain:
Not in our worst of nightmares could [we] have imagined that the drought would truly be of this
size and this proportion. (OrangeProc_A)
The novelty was the impact on processors and not the impact on farming, which is common,
though perhaps not at this scale; but we always have droughts and rainy periods. However,
processing plants have never suffered it [the impact]. (SugarAssoc)
Second– First-order Supply
Supply chain
order theme categories Position SC chain Quote resilience
Direct Productivity Farmer Sugarcane And as a result, we had a large loss in productivity
impact decrease – around 40% […] (SugarFarm)
Sugarcane Was there a decline in productivity? A lot – around
30% or more (Cachaça_D)
Gov. […] some sugar plantations lost 30% of their
production of sucrose and alcohol (GovAssist)
Processor Sugarcane So less sugarcane was grown, and there was less
productivity per hectare (SugarMill_B)
Sugarcane We ended 2014 […] with a shortfall (productivity)
of 17%, compared with what we planned
(Cachaça_Big)
Orange We are producing less than we produced
previously […] [less than] our potential
(OrangeProc_A)
Orange Yes, there were losses. There was a producer that
lost as much as 40% with the drought (ConCoop)
Orange It impacted our business here by more or less
R$ 150,000, some 5% [of revenues] (LocalCoop)
Orange In the crop it was productivity [impact] […] around
30% (OrangeProc_C)
Restricted Sugarcane We didn’t actually stop the mill, but we rationed
water supply [water] a lot. Our supply was very scarce
(SugarMill_V )
Sugarcane We started to use […] to treat water and reuse for
washing containers (CachaçaBig)
No Sugarcane We had very little impact […] from this water crisis
significant (Bioenergy)
impact Manufacturer Sugarcane/ What we noticed was that we had no impact […] in
Orange fact, we were favoured because my sales increased
[…] (F&B_MNE)
Sugarcane/ However, in general, we were not very affected […]
Orange (F&B_New)
Sugarcane/ The crisis in the operations did not upset us. It did
Orange not bring losses […] (F&B_Large)
Indirect Lower Processor Orange When the fruit came with production, the quality
impact quality was worse […] (OrangeProc_B)
Orange What happened was a reduction in fruit size
(LargeCoop)
Increased Processor Orange Therefore, the processor also had productivity
product cost problems and price problems for the product
(LargeCoop)
Manufacturer Sugarcane/ […] the price was absurd. The price of oranges Table III.
Orange increased by 70% […] (F&B_Small) Impact of the event

Second, organizations perceived and understood climate based on historical events, which
resulted in implicitly denying upcoming climate changes. Although respondents recognized
climate variations had increased in frequency and amplitude, those events were regarded as
“an exception.” The respondents had not reconciled historical perceptions of climate with
recent events; therefore, most were confident that there was no need to change traditional
procedures, such as choosing not to irrigate:
[…] there was a paradigm that, in the south of São Paulo, it was unnecessary to irrigate oranges
[…]. (OrangeProc_A)
[…] in our region the rainfall on average is very good […]. (SugarMill_I)
 IJOPM
level
adaptation – node
response and
Preparedness,
Table IV.
Second-order First-order
theme categories Position SC Supply chain Quote

Preparedness
Low Surprise Farmer Orange No. It was all very surprising (OrangeFarm)
awareness Sugarcane It was very frightening […] we’d never experienced a period as critical as that one (SugarFarm)
Processor Sugarcane Therefore, we started to become desperate […] “and now what’s the plan?” (SugarMill_B)
Orange […] we were caught by surprise, and we were unable to prepare (OrangeProc_A)
Climate Farmer Gov. You don’t irrigate sugarcane (GovAssist)
perception Sugarcane I think there’s a belief that this drought was an exception […] (Cachaça_C)
Processor Orange […] there was a paradigm that, south of São Paulo, it was unnecessary to irrigate oranges […]
(OrangeProc_A)
Sugarcane Drought only occurs sporadically […] in our region, the rainfall, on average, is very good […] (SugarMill_I)
Lack of Farmer Orange We get very little information […] (OrangeFarm)
information Processor Sugarcane We get weekly bulletins, and things don’t materialize […] (SugarMill_C)
Orange The producers’ access to information is a chronic problem we have in agriculture (LargeCoop)
Sugarcane This is a weakness we have because we truly need more information (SugarCoop)
Response
No response Nothing to do Farmer Sugarcane There was nothing we could do (SugarFarm)
Orange Because in fact we didn’t see many measures being taken (OrangeFarm)
Processor Sugarcane Therefore, you abort [production], and it’s no use. You have to live with it (SugarMill_E)
Orange […] there’s not much you can do. We just prayed to God to send rain (OrangeProc_C)
Flexibility Conserve Processor Orange […] in sugarcane, we have vinasse; we use vinasse to help, but it doesn’t solve the problem
water (OrangeProc_A)
Sugarcane In fact, we have too much water. We reuse it in some processes, and in others, there’s not an opportunity to
reuse it […] (Cachaça_Big)
Sugarcane As we have a lot of equipment with a closed circuit, we had no major problems (SugarMill_V )
Manufacturer Sugar/Orange I detailed internal projects to reduce waste (F&B_Large)
Redundancy Multi- Processor Orange Because we went after groves that had been irrigated (OrangeProc_B)
sourcing Sugarcane There are 11 mills in 5 different states, and so, it’s unlikely you’re going to have a catastrophe
(SugarMill_B)
Orange They normally buy from a wide region (LargeCoop)
Manufacturer Sugar/ Orange […] we weren’t affected because the company also has other mills in the northeast, mid-west, southeast
and south. (F&B_MNE)

(continued )
Second-order First-order
theme categories Position SC Supply chain Quote

No
Adaptation No Farmer/ Sugarcane […] if tomorrow we get another very severe crisis, I think we’re going to have problems (SugarMill_C)
adaptability prevention Processor
Orange At least from our side, we haven’t done anything effective yet (LargeCoop)
Gov. Regarding the sugarcane, we truly do depend on the rain (GovAssist)
Association […] the irrigated area has not been increasing as it should (Oranges Assoc)
Negative Farmer Sugarcane What could be done is you irrigate the sugarcane, but it was not economically feasible (Cachaça_A)
cost/benefit Sugarcane In our case, there’s not much to be done […] It’s not economically feasible to irrigate sugarcane
(Cachaça_D)
Sugarcane […] Irrigation of sugarcane is very expensive (SugarFarm)
Processor Orange You’re not going to irrigate; you’re not going to invest R$ 10,000 or R$ 12,000 per hectare to wait for a
weather event like this that might occur every 10 to 15 or 20 years (OrangeProc_A)
Sugarcane I think that [the problems are] the available technology and the cost, the investment (Cachaça_Big)
Willingness New Processor Sugarcane We decided in 2014 that we weren’t going to fail to make these investments [crop management]
to adapt investments (SugarMill_B)
Sugarcane […] we sought out these technical alternatives [water conservation] (SugarMill_Z)
Sugarcane We gradually increased our closed-circuit system (SugarMill_V )
Sugarcane Some decisions about varieties [in crops] were made because of their water resistance (SugarMill_I)
Manufacturer Sugar/ Orange We have plans to dig a new well to have a backup (F&B_New)

Supply chain
resilience
Table IV.
IJOPM Finally, these climate perceptions were corroborated by a lack of information regarding
climate change and extreme events. The main available weather sources provided, on
average, a 30-day forecast and did not anticipate long-term climate variance. Even
manufacturers acknowledged that there was a lack of awareness regarding the impacts of
climate change:
Look, we get very little information [climate]. (OrangeFarm)
4.2.2 The responses of firms. The way organizations responded to the event was grouped
into three different first-order categories: nothing to do, conserving water, and multi-
sourcing. Basically, reactions corresponded either to a no-response strategy or the use of
existent flexibility and redundancy to cope with the unexpected event (Table IV ). Given the
agricultural production cycle, farmers had no capacity to develop an immediate response to
the drought and lamented their losses. Most claimed that there was nothing to be done
regarding the drought:
The strategy was to pray a lot to see if it rains. (Cachaça_D)
Processors and manufacturers, in contrast, were able to react and respond to the drought
due to existing flexible processes. To manage the water shortage in the sugarcane mills and
the orange processing plants, processors accelerated the implementation of closed-circuit
and water conservation processes. Originally designed to comply with the Environmental
Protocol, the closed-circuit projects provided flexibility to processors in the reuse of water.
Manufacturers also analyzed how to reduce water consumption during their manufacturing
process and how to avoid wastewater in their facilities.
In addition to this flexibility, processors and manufacturers also responded to the event
by using multi-sourcing strategies to prevent indirect impacts. Integrated processors
accessed proprietary farms, which were geographically diversified. Non-integrated
processors and manufacturers searched for alternative providers in other regions to
ensure supply, given that the drought only affected the southeast region. In this sense, both
groups managed to mitigate the impact of the event with supply diversification and low
dependence on suppliers.
4.2.3 The adaptation for the future. During the interviews, firms were still in the process
of making sense of the past event, and recovery projects were not fully implemented.
Nevertheless, evidence of their intended adaptive strategies could be organized as follows:
no prevention, negative cost/benefit of projects and plans of new investments (Table IV ).
Some firms demonstrated no ability to adapt – mainly farmers and processors. A few
processors and manufacturers were willing to adapt, however, changing was not a priority
for most firms.
Regarding farming activities, respondents continued to view extreme events as rare and
not justifying an investment in adaptation measures. Some respondents recognized the need
to be prepared for future events, but they assessed the trade-off between costs and benefits
as negative and therefore preferred not to act. In both AFSCs, irrigation technologies were
considered too expensive and perceived as not economically viable, considering the
uncertainty of climatic events:
At least from our side, we have not done anything effective yet. (LargeCoop)
Therefore, even though it was a disaster, we thought that it was not worth investing because of the
cost [of the investment] and the return. (SugarMill_I)
Nevertheless, we could identify processors and manufacturers willing to invest in new
processes and strategies to mitigate the future occurrence of extreme weather events, such
as drilling new wells, crop management and water conservation projects. In the case of
manufacturers, aggravation of the crisis in major cities (including São Paulo) generated a Supply chain
sense of responsibility and public accountability in terms of water management. Concerned resilience
with their public image, manufacturers accelerated the development and the implementation
of eco-efficiency and water reuse projects; for them, the event was a catalyst for adaptation:
Some decisions about varieties [in crops] were made because of their water resistance. (SugarMill_I)
In summary, our data provided evidence that the nodes in the supply chain were affected in
distinct ways by the same event, resulting in a disturbance that had a limited effect on the
firms positioned downstream in the supply chain (Table V ). The vulnerability of farmers to
weather events confined them to a higher risk position in the supply chain. Farmers were
less prepared for the drought. With rigid configurations of processes and resources, they
were not able to respond to the event. Despite this fact, farmers continued to neglect the
climate risk and were reluctant to undertake any future action. Processors also
demonstrated low risk awareness but were able to respond to the event with flexible
processes and multi-sourcing strategies. The event also served as a catalyst for some
projects. Their investments in technologies and processes, however, depended greatly on the
cost-benefit ratio, which, for most processors, was still perceived as negative. Finally,
manufacturers were less impacted by the event but developed prompt responses to manage
supply risks. They also demonstrated a better understanding of the implications of climate
change and were willing to invest in new adaptation projects.

4.3 Supply chain findings

4.3.1 The supply chain’s preparedness. Our interviews provided evidence that there was no
dialogue among nodes and players in both supply chains. In fact, there was a general
perception of lack of information sharing among partners about the crisis, especially from
farmers and processors’ perspective (Table VI). Farmers and processors did not appear to
have a clear picture of the downstream supply chain and instead were mainly focused on
their own businesses. At the same time, the interviews did not provide any evidence
regarding a formal governance mechanism to promote the exchange of information about
the event in the supply chain. Even though information sharing and connectivity

Supply chain
(Orange and
Farmer Processor Manufacturer Sugarcane)

Event impact Direct impact Direct impact on farming units No significant direct No impact to final
on and restricted water supply impact; some impact on consumer
productivity (processing) the supply chain
Quality issues and price
increases due to impact to
farmers
Preparedness Low Low awareness Low awareness No visibility
awareness
Response No response Used flexibility and redundancy Used flexibility Few collaboration
capability to avoid major impacts and redundancy to initiatives in the
avoid impacts supply chain
Adaptation Reluctance to Reactive learning from the event. Willingness to adapt No evidence
undertake Some reluctance in undertaking with new investments
preventive preventive measures and some and environmental
measures willingness to act climate analysis Table V.
Resilience No Some Yes Yes Summary
IJOPM Second-order First-order Supply
theme categories Position SC chain Quote

Preparedness
No visibility No dialog Farmer Sugarcane No, I haven’t noticed anything like that
[dialog and collaboration] (Cachaça_C)
Processor Sugarcane No. It’s very difficult [dialog and
collaboration] (SugarMill_E)
No No support Manufacturer Sugar/ There was no specific action [with suppliers]
collaboration Orange because of a water shortage (F&B_New)
Other Association There was no direct support […] except in
terms of news/information [about the
drought] (SugarAssoc)
Response
Collaboration Cooperatives’ Processor Orange This irrigation department is specifically for
Support helping the cooperative members to overcome
this water shortage situation (LargeCoop)
Orange We’re training the cooperative members to
carry out some projects using irrigation
and changing crop management to prevent
this (ConCoop)
Collaboration Processor Orange We worked with the suppliers looking for
between buyer products and equipment that could help us
and supplier mitigate the risk in terms of a reduction in
spraying (OrangeProc_A)
Manufacturer Sugar/ We took part in some meetings and
Orange assessments with partners and suppliers,
[analyzing] what this might also mean in
terms of losses (F&B_Large)
Context
Lack of Fragmented Farmer Sugarcane Because the SugarAssoc is the industry union
cohesion sector(s) […] we’re midgets compared to them […]
(Sugar Farm)
Processor Sugarcane We’re a very fragmented industry
(Cachaça_Big)
Gov/ Gov. Everyone pulls in their own direction, even
Association the government bodies. No one works
together (GovAssist)
Association […] producers already know how to manage
their groves (Orange Assoc)
Sector Processor Orange Of the 17,000 citrus growers that existed 3
consolidation years ago, there are only 7,000 left (Orange
Proc_A)
Other Orange Before, there were 15,000 citrus growers,
and today, there are fewer than 2,000
(Orange Assoc)
Power Friction in SC Farmer Sugarcane The first thing is that the industry has to be
asymmetry more united (Sugar Farm)
Processor Sugarcane […] we should get these big companies together
and develop technologies (Sugar Mill_I)
Contracts Processor Orange This is very complicated in the industry.
You’d have to make contracts like the big
Table VI. firms, but we’re small (Orange Proc_C)
Preparedness, Manufacturer Sugar/ The small producers were affected,
response and context Orange particularly those that had no contracts or
– supply chain level short contracts (F&B_MNE)
throughout the supply chain are key to preparing the whole supply chain for any Supply chain
unexpected event, we found no evidence of visibility in the studied supply chains. resilience
4.3.2 The supply chain’s response. The supply chain responses depend on the collective
efforts of more than one node. Our findings revealed three different types of attitudes: no
support, cooperative support and vertical collaboration (Table VI). Most firms reported
there was no collaboration and that they were working in isolation with no joint action or
support from others. The interviews suggest only a few collective initiatives among the
nodes to manage the unexpected event, mainly restricted to orange cooperatives and
occasional initiatives:
The mills have that salvage irrigation because they work with the vinasse [a sugarcane by-product]
and we do not have access to that vinasse […] They say “but the vinasse is at your disposal”, but
then, we have to have the trucks, and we have to have everything, and all [collaboration] ends.
(SugarFarm)
In the orange supply chain, vertical collaboration was mainly identifiable in the support of
cooperatives for farmers. Cooperatives searched for partnerships with irrigation experts and
acted as important coordinators of farmers’ activities by helping farmers in terms of
information and training. Apart from those, coordinated initiatives in the two analyzed
supply chains were scarce, and two manufacturers and two processors promoted discussion
only with first-tier suppliers and partners to evaluate losses in the supply chain:
This irrigation department is specifically for helping cooperative members overcome this water
shortage situation. (LargeCoop)
Thus, our findings revealed that there was no collaboration among nodes, except for a few
cases downstream in the studied AFSCs and the vertical collaboration between farmers and
cooperatives in the orange supply chain.
4.3.3 The supply chain context. In terms of sector structure and relationship
interdependencies, four aspects of supply chain context emerged from our data analysis.
Specifically, both the fragmented nature of sectors and the process of sector consolidation
culminated in a lack of cohesion amongst the firms. Additionally, relationship friction and
governance based on ( formal) contracts demonstrated the power asymmetry in the two
analyzed supply chains (Table VI).
Both AFSCs have a larger number of players upstream than downstream – with only a few
manufacturing companies. Given the abundance of alternative players upstream, both supply
chains were regarded as highly fragmented. This structural aspect did not favor the
commitment of players to join forces; therefore, there was a lack of cohesion in the supply chains:
Everyone pulls in their own direction, even the government bodies. No one works together.
(GovAssist)
The non-cohesive context in both supply chains is related to a secondary contextual aspect:
the nature of the relationships. Relationship frictions demonstrated the conflict of interest
among players in orange and sugarcane supply chains. Additionally, the asymmetry in the
contractual relationships and the reliance on formal contracts configured the power
asymmetric context. Companies of both supply chains tended to manage their businesses
independently and have more formal governance mechanisms based on contracts:
If tomorrow for some reason […] you cannot produce, you will not be able to meet your contracts,
and this would be very damaging. (SugarMill_C)
In summary, our data analysis provides no evidence of some of the most cited elements in the
literature, such as velocity or adaptation, at the supply chain level. Similarly, the findings at
the supply chain level do not provide much evidence of visibility and collaboration,
IJOPM suggesting that resilience was primarily developed at the firm level. In terms of supply chain
context, both supply chains can be regarded as highly fragmented and with few interactions
among actors within and across nodes. In the next section, we discuss our findings vis-à-vis
the literature based on the emergent model proposed in Figure 3 by comparing the resilience
built at each node within the overall SCRES.

5. Discussion
In this section, we explain the relational dynamic among the aggregate dimensions that
emerged from our findings to answer our research question:
RQ1. How does resilience, built at different nodes of a supply chain, influence SCRES?
Literature suggests that to cope with and recover from the impacts of unexpected events,
organizations need to build different resilience elements at the individual and supply chain
level, and at each phase of the disruption (Hohenstein et al., 2015; Ali et al., 2017; Stone and
Rahimifard, 2018).
Figure 3 illustrates the elements found in our study at each resilience phase and the
interaction with the contextual aspects of those supply chains. While lack of awareness,
responsiveness and adaptation capability was traits identified at upstream nodes, SCRES was
mainly supported by the flexibility and redundancy of processors and manufacturers, instead of
visibility and collaboration in the supply chain. The occasional collaborative initiatives were
related to adaptation plans; however, not all adaptation plans were based on collaboration.
Therefore, our study evidences that resilience elements were primarily built at the node level,
not at the supply chain level. In the investigated supply chains, firms sought isolated solutions
rather than using common strategies to cope with the risk for the whole supply chain.
Our research suggests that the preference for individual strategies, rather than supply
chain solutions, is related to a lack of cohesion and asymmetry in the supply chains. Thus,
the interfirm relationship context is important to the development of resilience at firms and
the supply chain levels, as explained herein.
Resilience
Phases

Preparedness Response Adaptation

Low No response No adaptability

awareness (upstream) (upstream)
Node level
Resilience Elements

Flexibility Redundancy Willingness to

(downstream) (downstream) adapt
(downstream)

Collaboration
SC level

Low
visibility
No
collaboration
Supply Chain

Figure 3.
Context

Lack of Power
Emergent model of cohesion asymmetry
nodes and supply
chain dynamics
5.1 Resilience at the node level and SCRES Supply chain
We found no evidence of significant impact or disruption at the end of the supply chain. resilience
Considering that SCRES is the capacity of the supply chain to maintain operational
continuity and satisfy the end consumer (Ponomarov and Holcomb, 2009; Kamalahmadi and
Parast, 2016; Stone and Rahimifard, 2018), our findings reveal that both supply chains were
resilient, despite the evidence of no resilience at farmers’ node.
However, the compared analysis demonstrates different levels of impact as well as
different degrees of preparedness, response and adaptation among the three nodes (Table V ).
Farmers, the most vulnerable node to the climatic events, were the least resilient, with no
evidence of elements to anticipate, respond, or even to learn from the event. As we moved
downstream in the AFSC, we identified different resilience elements adopted by processors
and manufacturers, such as flexibility and redundancy, that were developed in advance, and
allowed them to respond and control the disturbance and fulfill the demands of end
consumers. Our findings highlighted that their resilience strategies were not dependent on the
development of resilience at previous stages in the supply chain. Instead, SCRES depended on
the capacity of downstream organizations to manage unexpected events. Our study
demonstrates that SCRES is not dependent on the resilience of each node, as suggested in the
literature (Leat and Revoredo-Giha, 2013; Stone and Rahimifard, 2018) and that different levels
of resilience in a supply chain may not jeopardize SCRES. Thus:
P1. SCRES is not dependent on the resilience of all firms in the SC.
This proposition also suggests that SCRES may not require resilience in the overall system
but rather specifically from some key players. Most studies emphasize the perspective of a
focal company in the analysis of supply chain and thus overlook the impacts, response and
recovery at some nodes (Leat and Revoredo-Giha, 2013; Tukamuhabwa et al., 2015;
Birkie et al., 2017). Our approach allows the observation of the role of different nodes in the
continuity of the supply chain operations and reveals that SCRES can be achieved even
when upstream organizations are not resilient. In this sense, it is important to understand
that the concept of SCRES does not apply to all organizations in the supply chain but
is primarily associated with the company that is able to reconfigure the resources and
players to control the disruption. Therefore, the analysis of the nodes offered a different
perspective of SCRES.

5.2 Resilience elements

SCRES depends on both firm and SCRES elements that should be developed prior to, during
and after the disruption (Hohenstein et al., 2015; Ali et al., 2017; Stone and Rahimifard, 2018).
Organizations may adopt different resilience strategies depending on the risk source
(Tukamuhabwa et al., 2015) and the supply chain context and environment (Birkie et al.,
2017; Stone and Rahimifard, 2018). Nevertheless, the empirical studies on SCRES still have
to analyze the relationship among the resilience elements (Tukamuhabwa et al., 2015;
Stone and Rahimifard, 2018). By investigating different nodes in the supply chain, our
research brings new insights into how the supply chain context influences the adoption of
different elements to cope with vulnerabilities.
The data analysis revealed a context of low connectivity and contractual relationships
among firms that did not foster cohesion and joint efforts to manage the disruption
(Lawler and Yoon, 1996; Fiksel, 2003). In the environment of relative independence and
asymmetric relationships between buyer–suppliers, companies focus on their own interests
rather than on common goals, and there is no great need for coordination of activities
between partners. Consequently, firms make individual decisions in response to any
changes in their environments (Pathak et al., 2007). Therefore, when facing unexpected
events, stronger firms will hedge their risks through redundant capacity and independent
IJOPM resources (Caniëls and Gelderman, 2007; Jüttner and Maklan, 2011). In this sense, the lack of
cohesion in both supply chains investigated and the power asymmetry among the nodes
seemed to drive the resilience strategies towards firm-level elements, such as flexibility and
redundancy, rather than supply chain elements, such as collaboration. Our results
corroborate with those of Stone and Rahimifard (2018) in that resilience elements are
enabled by different variables at the node and the supply chain level. Therefore:
P2. In a supply chain context of low interdependence, resilience elements at the node
level will be preferred to those at the supply chain level.
This proposition highlights the importance for managers to understand their specific
supply chain context in order to better allocate resources and invest in the most
appropriate resilience elements to reduce vulnerabilities (Fiksel et al., 2015). Nevertheless,
the choice of node level solutions might not result in the best overall performance
for the supply chain. For example, although the multi-sourcing strategy was successful
for processors and manufacturers, it did not provide support for the development
of the farming stage. Consequently, farmers – with no information, no visibility and no
support – were not willing to adapt to the new climatic scenario. The current process
forces downstream nodes to search for individual solutions to assure the continuity of
their operations and might have a negative effect on SCRES. Alternatively, if the whole
supply chain is involved in building resilience elements for each phase, results in terms of
time and cost might be optimized.
System solutions, rather than end-user focus, are crucial in networks such as AFSC
(Stone and Rahimifard, 2018). Interfirm ties are important for adaptation and learning;
therefore, a lack of collaboration in a supply chain does not promote the development of
adaptation capabilities (Zollo et al., 2016) and represents a high risk to AFSC due to the
cyclical and cumulative nature of the resilience building process (Stone and Rahimifard, 2018).

6. Conclusions
This research investigated how resilience, built at different nodes of a supply chain,
influences overall SCRES, bringing a new perspective to the literature that normally focuses
on the analysis of a focal company (Wieland and Wallenburg, 2013; Brandon-Jones et al.,
2014; Tukamuhabwa et al., 2015). In this sense, this research contributes to knowledge about
SCRES in two ways.
First, it provides a broader perspective of resilience in the supply chain by analyzing and
comparing the impact, preparedness, response and adaptation at each supply chain node.
The analysis of different perspectives ( farmers, processors and manufacturers)
demonstrates an independence of firms in the supply chain and the different relevance of
firms when securing SCRES. Moreover, it suggests that SCRES can be primarily associated
with a resilient organization downstream in the supply chain.
Second, this study compares node and SCRES elements in a supply chain context
with low interdependence among firms. In this case, node resilience elements are
preferred to those at the supply chain level. Therefore, this study also sheds light on the
influence of the supply chain context (interfirm relationships) in the process of developing
resilience strategy.

6.1 Managerial and social implications

This study highlights the interdependence of firms’ resilience and SCRES, and the
importance of individual resilience elements in certain supply chain contexts. For managers,
our findings suggest the need to evaluate vulnerabilities and alternative resilience elements
at different supply chain nodes to assure SCRES.
 For more vulnerable nodes, the findings suggest that exchange of information and Supply chain
collaboration in the supply chain have to be built prior to disruptive events. Specifically, in resilience
AFSC, there is a need to intensify the dissemination of climate information, particularly
among farmers, to increase their risk awareness and willingness to adapt. As suggested in
this study, associations, cooperatives and government organizations are relevant
in fostering the knowledge and training in upstream AFSC. They may also promote
knowledge exchange and collaboration in the supply chain. This aspect is key to increasing
the overall SCRES.
The research indicates that, for firms located downstream in the supply chain, current
choices of resilience strategy have been able to assure continuity of operations but fail to
address future uncertainties. In scenarios with an increasing frequency of climate extremes,
current solutions may not be sufficient. Particularly in AFSC, downstream companies could
help upstream companies build resilience in order to prepare for climatic events, such as
slow-onset droughts that are cumulative and difficult to perceive immediately. Developing
SCRES elements rather than organizational ones might improve the SCRES.

6.2 Limitations and further research

Our study has some limitations. The first limitation refers to the case studies. Although our
research contained firms from two different AFSC, the agribusiness has peculiarities in
terms of climate dependence and market environment that are not present in all industries.
Our study was performed in a single institutional context of an emerging economy, with
implications for the dynamics of interfirm relationships. The second limitation relates to the
case study method and the difficulty in establishing the external validity of the study
findings (Eisenhardt, 1989). Triangulation with other documents and interviews with
associations and influential stakeholders were intended to reduce this problem, but the
findings of this study should be tested in different institutional and supply chain contexts
to validate the findings. Further research should also explore the difference between
slow-onset and sudden-onset events.

References
Ali, A., Mahfouz, A. and Arisha, A. (2017), “Analysing supply chain resilience: integrating the
constructs in a concept mapping framework via a systematic literature review”, Supply Chain
Management: An International Journal, Vol. 22 No. 1, pp. 16-39.
Ambulkar, S., Blackhurst, J. and Grawe, S. (2015), “Firm’s resilience to supply chain disruptions: scale
development and empirical examination”, Journal of Operations Management, Vols 33-34 No. 1,
pp. 111-122.
Barratt, M. and Oke, A. (2007), “Antecedents of supply chain visibility in retail supply chains: a
resource-based theory perspective”, Journal of Operations Management, Vol. 25 No. 6,
pp. 1217-1233.
Barratt, M., Choi, T.Y. and Li, M. (2011), “Qualitative case studies in operations management: trends,
research outcomes, and future research implications”, Journal of Operations Management,
Vol. 29 No. 4, pp. 329-342.
Bhamra, R., Dani, S. and Burnard, K. (2011), “Resilience: the concept, a literature review and future
directions”, International Journal of Production Research, Vol. 49 No. 18, pp. 5375-5393.
Birkie, S.E., Trucco, P. and Fernandez Campos, P. (2017), “Effectiveness of resilience capabilities in
mitigating disruptions: leveraging on supply chain structural complexity”, Supply Chain
Management: An International Journal, Vol. 22 No. 6, pp. 506-521.
Brandon-Jones, E., Squire, B., Autry, C.W. and Petersen, K.J. (2014), “A contingent resource-based
perspective of supply chain resilience and robustness”, Journal of Supply Chain Management,
Vol. 50 No. 3, pp. 55-73.
IJOPM Burnard, K., Bhamra, R. and Tsinopoulos, C. (2018), “Building organisational resilience: four
configurations”, IEEE Transactions on Engineering Management, Vol. 65 No. 3, pp. 351-362,
available at: https://doi.org/10.1109/tem.2018.2796181 (accessed May 4, 2018).
Caniëls, M.C.J. and Gelderman, C.J. (2007), “Power and interdependence in buyer supplier relationships:
a purchasing portfolio approach”, Industrial Marketing Management, Vol. 36 No. 2, pp. 219-229.
Cao, M. and Zhang, Q. (2011), “Supply chain collaboration: impact on collaborative advantage and firm
performance”, Journal of Operations Management, Vol. 29 No. 3, pp. 163-180.
Christopher, M. and Peck, H. (2004), “Building the resilience supply chain”, International Journal of
Logistics Management, Vol. 15 No. 2, pp. 1-13.
Eisenhardt, K.M. (1989), “Building theories from case study research”, The Academy of Management
Review, Vol. 14 No. 4, pp. 532-550.
Eisenhardt, K.M. and Graebner, M.E. (2007), “Theory building from cases: opportunities and challenges”,
Academy of Management Journal, Vol. 50 No. 1, pp. 25-32.
Fiksel, J. (2003), “Designing resilient, sustainable systems”, Environmental Science and Technology,
Vol. 37 No. 23, pp. 5330-5339.
Fiksel, J., Polyviou, M., Croxton, K.L. and Pettit, T.J. (2015), “From risk to resilience: learning to deal
with disruption”, MIT Sloan Management Review, Vol. 56 No. 2, pp. 79-86.
Flint, D.J., Woodruff, R.B. and Gardial, S.F. (2002), “Exploring the phenomenon of customers’ desired
value change in a business-to-business context”, Journal of Marketing, Vol. 66 No. 4, pp. 102-117.
Gioia, D.A., Corley, K.G. and Hamilton, A.L. (2013), “Seeking qualitative rigor in inductive research:
notes on the Gioia methodology”, Organizational Research Methods, Vol. 16 No. 1, pp. 15-31.
Hohenstein, N.O., Feisel, E., Hartmann, E. and Giunipero, L. (2015), “Research on the phenomenon of
supply chain resilience: a systematic review and paths for further investigation”, International
Journal of Physical Distribution & Logistics Management, Vol. 45 Nos 1/2, pp. 90-117.
Jüttner, U. (2005), “Supply chain risk management: understanding the business requirements from a
practitioner perspective”, The International Journal of Logistics Management, Vol. 16 No. 1,
pp. 120-141.
Jüttner, U. and Maklan, S. (2011), “Supply chain resilience in the global financial crisis: an empirical
study”, Supply Chain Management: An International Journal, Vol. 16 No. 4, pp. 246-259.
Jüttner, U., Peck, H. and Christopher, M. (2003), “Supply chain risk management: outlining an agenda
for future research”, International Journal of Logistics: Research & Application, Vol. 6 No. 4,
pp. 197-210.
Kamalahmadi, M. and Parast, M.M. (2016), “A review of the literature on the principles of enterprise
and supply chain resilience: major findings and directions for future research”, International
Journal of Production Economics, Vol. 171 No. 1, pp. 116-133.
Kaufmann, L. and Denk, N. (2011), “How to demonstrate rigor when presenting grounded theory
research in the supply chain management literature”, Journal of Supply Chain Management,
Vol. 47 No. 4, pp. 64-72.
Kaufmann, L., Esslinger, J. and Carter, C.R. (2018), “Toward relationship resilience: managing buyer-
induced breaches of psychological contracts during joint buyer–supplier projects”, Journal of
Supply Chain Management, Vol. 54 No. 4, pp. 62-85.
Lawler, E.J. and Yoon, J. (1996), “Commitment in exchange relations: test of a theory of relational
cohesion”, American Sociological Review, Vol. 61 No. 1, pp. 89-108.
Leat, P. and Revoredo-Giha, C. (2013), “Risk and resilience in agri-food supply chains: the case of the
ASDA PorkLink supply chain in Scotland”, Supply Chain Management: An International
Journal, Vol. 18 No. 2, pp. 219-231.
Liu, Y., Luo, Y. and Liu, T. (2009), “Governing buyer–supplier relationships through transactional and
relational mechanisms: evidence from China”, Journal of Operations Management, Vol. 27 No. 4,
pp. 294-309.
Maloni, M.J. and Brown, M.E. (2006), “Corporate social responsibility in the supply chain: an application Supply chain
in the food industry”, Journal of Business Ethics, Vol. 68 No. 1, pp. 35-52. resilience
Mosqueira-Salazar, K.J., Palacios-Bereche, R., Chávez-Rodríguez, M., Seabra, J. and Nebra, S.A. (2013),
“Reduction of water consumption in an integrated first- and second-generation ethanol plant”,
Energy for Sustainable Development, Vol. 17 No. 5, pp. 531-535.
Neves, M.F., Trombin, V.G. and Kalaki, R.B. (2013), “Competitiveness of the orange juice chain in
Brazil”, International Food and Agribusiness Management Review, Vol. 16 No. 4, pp. 141-158.
Pathak, S.D., Day, J.M., Nair, A., Sawaya, W.J. and Kristal, M.M. (2007), “Complexity and adaptivity in
supply networks: building supply network theory using a complex adaptive systems
perspective”, Decision Sciences, Vol. 38 No. 4, pp. 547-580.
Pettit, T.J., Croxton, K.L. and Fiksel, J. (2013), “Ensuring supply chain resilience: development and
implementation of an assessment tool”, Journal of Business Logistics, Vol. 34 No. 1, pp. 46-76.
Ponis, S.T. and Koronis, E. (2012), “Supply chain resilience: definition of concept and its formative
elements”, Journal of Applied Business Research, Vol. 28 No. 5, pp. 921-930.
Ponomarov, S.Y. and Holcomb, M.C. (2009), “Understanding the concept of supply chain resilience”,
The International Journal of Logistics Management, Vol. 20 No. 1, pp. 124-143.
Revilla, E. and Saenz, M.J. (2017), “The impact of risk management on the frequency of supply chain
disruptions: a configurational approach”, International Journal of Operations and Production
Management, Vol. 37 No. 5, pp. 557-576.
Rice, J.B. and Caniato, F. (2003), “Building a secure and resilient supply network”, Supply Chain
Management Review, Vol. 7 No. 5, pp. 22-30.
Roth, A.V., Tsay, A.A., Pullmand, M.E. and Gray, J.V. (2008), “Unraveling the food supply chain:
strategic insights from China and the 2007 recalls”, Journal of Supply Chain Management, Vol. 44
No. 1, pp. 22-39.
Scholten, K. and Schilder, S. (2015), “The role of collaboration in supply chain resilience”, Supply Chain
Management: An International Journal, Vol. 20 No. 4, pp. 471-484.
Sheffi, Y. and Rice, J.B. Jr (2005), “A supply chain view of the resilient enterprise”, MIT Sloan
Management Review, Vol. 47 No. 1, pp. 41-48.
Stevenson, M. and Spring, M. (2007), “Flexibility from a supply chain perspective: definition
and review”, International Journal of Operations & Production Management, Vol. 27 No. 7,
pp. 685-713.
Stone, J. and Rahimifard, S. (2018), “Resilience in agri-food supply chains: a critical analysis of the
literature and synthesis of a novel framework”, Supply Chain Management: An International
Journal, Vol. 23 No. 3, pp. 207-238.
Strauss, A.L. and Corbin, J. (1998), Basics of Qualitative Research: Techniques and Procedures for
Developing Grounded Theory, Vol. 2, Sage, Thousand Oaks, CA.
Sullivan-Taylor, B. and Branicki, L. (2011), “Creating resilient SMEs: why one size might not fit all”,
International Journal of Production Research, Vol. 49 No. 18, pp. 5565-5579.
Tukamuhabwa, B.R., Stevenson, M., Busby, J. and Zorzini, M. (2015), “Supply chain resilience:
definition, review and theoretical foundations for further study”, International Journal of
Production Research, Vol. 53 No. 18, pp. 5592-5623.
Tukamuhabwa, B.R., Stevenson, M. and Busby, J. (2017), “Supply chain resilience in a developing
country context: a case study on the interconnectedness of threats, strategies and outcomes”,
Supply Chain Management: An International Journal, Vol. 22 No. 6, pp. 485-505.
UNICA (2015), “Sugarcane, one plant, many solutions – sugar, ethanol, bioelectricity & beyond”,
available at: http://sugarcane.org/resource-library/books/Folder%20and%20Brochure.pdf
(accessed January 1, 2017).
Van Der Vegt, G.S., Essens, P., Wahlström, M. and George, G. (2015), “Managing risk and resilience”,
Academy of Management Journal, Vol. 58 No. 4, pp. 971-980.
IJOPM Wieland, A. and Wallenburg, C.M. (2013), “The influence of relational competencies on supply chain
resilience: a relational view”, International Journal of Physical Distribution & Logistics Management,
Vol. 43 No. 4, pp. 300-320.
Yin, R.K. (1989), Case Study Research: Design and Methods, Sage publications, Newbury Park, CA.
Zollo, M., Bettinazzi, E.L., Neumann, K. and Snoeren, P. (2016), “Toward a comprehensive model of
organizational evolution: dynamic capabilities for innovation and adaptation of the enterprise
model”, Global Strategy Journal, Vol. 6 No. 3, pp. 225-244.

Further reading
Carvalho, H., Azevedo, S.G. and Machado, V.C. (2014), “Supply chain management resilience: a theory
building approach”, International Journal of Supply Chain and Operations Resilience, Vol. 1 No. 1,
pp. 3-27.

Appendix – Interviews Protocol

(1) Before this extreme drought, have you ever faced a similar event? How often do these severe
droughts affect the company?
(2) Considering your experience in agri-food supply chain and climate. Was this event different
from others? In what way?
(3) How did this extreme drought affect your company? Can you give us some examples?
(4) How did you know that something new was going on?
(5) From the moment you perceived this was an extreme event, what measures were adopted by
the organization related to your processes? Were there changes adopted by the organization in
terms of resources? Can you explain which measures were more effective?
(6) During the drought, did you work with buyers/suppliers to better respond to it? Can you give
us any example?
(7) How did the actions taken by buyers/suppliers affect your reaction to the drought?
(8) Can you give us any example of how do you monitor such climate events? How reliable
is the information available today? Can you tell us about your source of information about
weather conditions?
(9) Can you tell us if you have plans for future events like this one? Do you have plans for new
investments to avoid losses in future events?

Corresponding author
Priscila Laczynski de Souza Miguel can be contacted at: [email protected]

For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: [email protected]