Why are some firms more resilient than others? The effect of firm and managerial characteristics on organizational resilience

(1)

Why are some firms more resilient than others?

The effect of firm and managerial characteristics on

organizational resilience

MSc International Business & Management

Faculty of Economics and Business

University of Groningen

Steven Künzli Student number: 2003368 s.g.kunzli@student.rug.nl Supervisor: Dr. G. de Jong 16th of January, 2016

(2)

2 ABSTRACT

This thesis extends the literature by providing a start in the development of a ‘’grant theory’’ about organizational resilience. Besides an extensive literature review in various domains, we tested the impact of firm age, firm size, managerial education and managerial experience on organizational resilience. A survey was conducted under 4500 small and medium enterprises in the northern part of the Netherlands (10,6% response rate) and an Ordinary Least Squares (OLS) was used to test the various relations.

Findings indicated that the used firm characteristics did not have any significant relation with organizational resilience. Personal characteristics showed mixed results: after univariate analysis, managerial education showed significant and negative results in relation to organizational resilience. Managerial experience showed a significant and positive relation with organizational resilience after univariate analysis and this significance still proven when combined with managerial education.

In conclusion, this thesis conducts a research to the effects of firm and managerial characteristics on organizational resilience and provides a solid base for future empirical research.

(3)

3

3.3 Operationalizing variables ... 37 3.3.1 Dependent variable ... 38 3.3.2 Independent variable ... 38 3.3.3 Control variable ... 39 3.4 Statistical model ... 41 3.5 Method assumption ... 42 3.5.1 Homoscedasticity ... 43 3.5.2 Multicollinearity ... 43 3.5.3 Normality ... 44 4. Empirical Results ... 45 4.1 Descriptive results ... 45 4.2 Regression results ... 46

4.3 Discussion of the results ... 49

5. Robustness tests ... 51

(4)

4

5.2 Disaggregating Organizational Resilience ... 52

5.3 Different estimation method ... 53

6. Conclusion ... 55

6.1 Added value ... 56

6.1.1 Theoretical contribution ... 56

6.1.2 Managerial implications ... 57

6.2 Limitations and future research ... 58

7. Appendix ... 61

7.1 Appendix 1 (Survey) ... 61

7.2 Appendix 2 (Operationalization variables)... 63

7.3 Appendix 3 (Method Assumptions) ... 64

7.4 Appendix 4 (Empirical Results) ... 67

7.5 Appendix 5 (Robustness tests) ... 69

(5)

5

1. Introduction

The environment demands organizations to continually adapt in order to remain viable and to sustain competitiveness (Burnard and Bhamra, 2011). Their environment is characterized by disruptions and discontinuities on various scales, which include disasters like earthquakes, hurricanes and terrorist attacks but also small uncertainties or deviations such as human errors.

In the beginning of 21st century there was a feeling of contentment in the world. One thought that most societies and economies were developing steadily and the biggest threat was the Millennium bug. Furthermore, the occurrence of crisis and recessions were seen as unlikely because it was thought that politicians, economists and business leaders mastered the way how societies, economies and businesses should be managed. The credit crunch of 2008 changed everything completely. The developed world was thrown into a deep recession that appeared to be one of worst of the last century. Many companies did not survive this recession since they were unable to adapt to the changed economic landscape. Others did survive because they were able to notice the change, the way it was going, they understood the consequences and were able to react to these changes in a appropriate way (Strycharczyk and Elvin, 2014). The businesses that survived showed high levels of resilience during this global financial crisis, which is a perfect example of a disruptive event of major proportions.

Businesses all over the world needed to cope with different kinds of disruptions on various scales. Therefore, the resilience construct could be applied on a global scale.

Resilience is a concept, which is used in a variety of domains like fibers and fabrics, ecology, and psychology. Although the context of the definition may change, the term resilience is closely related to the ability of an component to return to an equilibrium state after a disruption across all fields. Holling (1973 in Bhamra et al., 2011, p. 5380), who first started to popularize the concept, argued that resilience ‘‘is a measure of systems persistence and the ability to absorb disturbances and still maintain the same relationships between system entities.’’

(6)

6

building but literature which focuses on empirical research (i.e. what makes certain companies more resilient than others) is still scarce. Therefore, this thesis will conduct an empirical research to organizational resilience.

Organizational resilience relates to both individual and organizational responses to disruptions (Bhamra et al. 2011). Many scholars (such as Lengnick-Hall et al., 2011, Coutu, 2002, Bannano & Galea, 2007) studied individual resilience and argued that resilience is a combination of nature and nurture. They concluded that some people are more resilient than others. Bannano and Galea (2007) attributed personal resilience to a variety of characteristics (e.g. age, race, education, level of trauma exposure, income level), while Couto (2002) argued that there are just three defining characteristics of resilient people: they calmly accept the severe realities that are facing them, they search for meaning in dreadful times, and they continuously improvise. Although individuals are part of organizations and therefore have a direct influence on a firms capacity for resilience ( Lengnick-Hall et al., 2011), concepts of individual resilience cannot be generalized to organizational resilience. Organizations are composed of a complex network of subsystems (containing technical as well as social components) and interrelated elements that interact through nonlinear relationships (Burnard & Bhamra. 2011). Other factors like organizational characteristics, operational environment and resource availability influences organizational resilience as well (Lee et al., 2013).

(7)

7

firm size increases. This will help us explain why some firms are more resilient than others. The research question is of this thesis is:

Why are some firms more resilient than others?

The following sub questions will be addressed to answer the research question  What is organizational resilience and how can it be measured?

 What is the relationship between firm size and organizational resilience?  Is there a relationship between firm age and organizational resilience?

 Do differences in educational backgrounds of managers influence organizational resilience?

 Does managerial experience influence organizational resilience?

 Does firm size impact the educational effectiveness of a manager in improving organizational resilience?

Besides a literature review, this thesis includes a quantitative empirical research on organizational resilience in the northern part of the Netherlands. This thesis chose this setting for multiple reasons that will be explained in the methods section. The goal of the thesis is to underline the importance of organizational resilience and find factors that make firms more resilient than others. Due to the fact that the main research question is of a general nature, this thesis will focus on the effects of various independent variables on the level of organizational resilience. Existing organizational performance, innovation and change literature will help construct the proposed hypothesis as well as existing resilience literature. Data was gathered by sending out cross-sectional surveys to various organizations. This survey will capture the level of organizational resilience and important characteristics of the firms and managers involved.

(8)

8

2. Theoretical Background

This section will start with an overview of resilience literature in the domains of fibers and fabrics, ecology, psychology and economy. These domains provided the base of organizational resilience. Then, we will continue with the current business literature on resilience and how we can measure the construct. We will conclude this section by discussing the formulated hypothesis.

2.1 Origin of resilience

The concept of resilience can be found in literature of various domains. Although there are generally used definitions for resilience, they are also discipline specific definitions. However, in many cases the resilience construct is often ambiguous (Ponomarov and Holcomb, 2009). The critiques about the resilience concept are mainly focused on ambiguities in definitions, instability of the occurrence of resilience, and the usefulness of resilience as a theoretical construct (Luthar et al., 2000). Therefore, to fully understand the concept of resilience, this thesis constructs a chronological literature review in different streams of literature and considers various perspectives and approaches. This way, the thesis directs the reader through the development of the resilience concept and identifies the common features. This section will elaborate more about ecological and psychological resilience due to the fact that those areas are more relevant for organizational resilience.

2.1.1 Fibers and Fabrics

(9)

9

This definition is valuable because it provides a starting point for distinguishing between the extent and kind of resiliency. The extent of resilience is the degree of recovery in a certain timeframe. The resilience coefficient of two materials could be equal but this does not mean that they are the same. There are more differences between the two materials besides this coefficient. Therefore, four kinds of resilience can be distinguished, which have either high or low moduli and high or low rates of recovery. Furthermore, there is a difference between resiliency and resilience according to the fibers and fabrics literature; ‘‘resiliency is a specific property of a substance, while resilience is a total property depending not only on the material but on the size and shape’’ (Hofmann, 1948, p. 148). Resilience in materials can vary in different types of the same materials (Demiruren and Burns, 1955). Dillon (1947) developed a resilience ratio that measures the energy of retraction against the energy of deformation. He concluded that the ratio is influenced by several factors (Dillon, 1947, p. 209):

1. Temperature 2. Moisture content 3. Rate of strain 4. Rate of retraction 5. Strain history

6. Rest period between cycles 7. Nature of cycles

a. Constant rate of strain b. Constant rate of loading c. Constant maximum strain d. Constant maximum stress e. Periodic

f. Impact

(10)

10

term resilience in fibers and fabrics provides a fundamental molecular explanation of the physical and mechanical properties. (Clark et al., 1940) and is seen as the start of the resilience literature. The next section will dive into the ecology literature and takes systems into account.

2.1.2 Ecology

This section discusses ecological resilience, which could be seen as the basis for organizational resilience. In 1973, Holling started to popularize the term resilience within his seminal paper ‘Resilience and Stability of Ecological systems’. This article is seen as the most important reference for the majority of studies related to the concept of resilience from that moment on. Before elaborating more about resilience in the ecological context, this section will explain the importance of systems in the construct.

(11)

11

Figure 1: Ecosystem functions

Source: Holling (1987)

Holling (1973) introduced the viewpoints ‘resilience’ and ‘stability’ of ecosystems in his earlier work. Holling defined resilience as the ability of a system to absorb the different kind of changes, while stability refers to the ability to return to an equilibrium state after a disruption. Clapham (1971) agreed and argued that stability creates a pre-disturbance state instead of just an adjustment to a new equilibrium state that could be better or worse than the previous one. While Gallopin (2006) showed that ecological resilience does not necessarily mean that a system returns to the state it was before a disruption. According to him, ecological resilience can be measured by ‘‘the magnitude of the perturbation that can be absorbed before the state of the system falls outside its domain of attraction (Gallopin, 2006, p.299)’’. Stability, also called engineering resilience (Holling, 1996), can be measured by the speed it returns to its equilibrium or stable point.

The concept of resilience evolved since Holling wrote his paper. Westman (1986 in Ponomarov and Holcomb, 2009) defined several components of resilience, which lead to a more detailed definition of resilience in ecological literature. He defined resilience as ‘‘the degree, manner and pace of restoration of initial structure and function in an ecosystem after disturbance (Westman, 1978; Vlapham, 1971 in Ponomarov and Holcomb, 2009, p. 26)’’. The components that Westman associated with resilience are:

 Elasticity: The speed in which a system returns to a stable state.

 Amplitude: The zone of distortion from which the system return to its equilibrium  Hysteresis: The extent to which the path from and back to the stable state are not

(12)

12  Malleability: The degree to which the equilibrium state after the disruption differs

from the original state.

 Damping: The degree and manner by which the path of repair is influenced by any force that changes the normal path of restoration.

Inertia should be considered as well, according to Westman (1986). Inertia is the resistance of a system when it perceives stress due to changing circumstances. If inertia is high, it is harder for a system to be resilient. Each of the five above-mentioned components will reveal a different level of resilience to a certain disturbance. It is hard to measure the degree of recovery because of speculative events and sample variability. However, in cyclically stable systems, short term (within cycles) and long term (between cycle) recovery can be measured by periodicity (Westman, 1986).

Carpenter et al. (2001) introduced the adaptive cycle theory, which states that dynamic systems do not turn to equilibrium but they evolve through four states. These states are (1) rapid growth and exploitation (2) preservation, (3) creative destruction and (4) renewal. He concluded that resilience has 3 important characteristics. First, resilient systems can retain the same controls on functions and structure in time of change. Second, resilient systems are able to organize themselves without interference from external factors. Finally, resilient systems have the capacity to learn and adapt to disruptions.

Elmqvist et al. (2003) also highlighted the importance of adaptive capacity for resilient ecosystems. Adaptive capacity can be defined as ‘‘the ability of a system to adjust to climate change (including climate variability and extremes), to moderate potential damages, to take advantage of opportunities, or to cope with the consequences (Brooks et al., 2005, p. 153)’’. Important in creating resilience is the diversity in response to environmental change among species that contribute to the same ecosystem function, especially during system renewal and reorganization after change. Furthermore, it provides adaptive capacity in a world of uncertainty, human-dominated environments and complex systems (Elmqvist et al., 2003).. Adaptive capacity is also an important pillar in organizational resilience but this thesis will elaborate more on this subject later on in this chapter

(13)

13

external stresses and disturbances as a result of social, political, and environmental change (Adger, 2000 in Gallopin 2006, p.297)’’. Hughes et all (2005) argued that it is much easier to preserve ecosystems against disturbances than to restore them after they are tarnished. Therefore, they conclude that adaptive governance systems are also needed in socio-ecological perspective to protect natural resources, which are too valuable, economically, aesthetically and culturally. Even without disruptions in their environment, people could become better adapted by improving their condition in it (Gallopin, 2006). The socio-ecological literature also highlights the importance of adaptive capacity. Nelson (2007) argues that adaptive capacity can be seen as a core feature of the socio-ecological resilient approach. This approach is system oriented and has a dynamic view because it links resources to environmental outcomes. Adaptive capacity could also create new stability domain or system equilibriums (Bhamra et al., 2011). Resilient systems are able to withstand surprises, renew the system, reduce vulnerabilities of specific groups to an acceptable level, create robustness to uncertainty and maintain flexibility. Furthermore, they also need to recognize if a disruption has got minor, short-term consequences, or major consequences (Nelson et al., 2007). As you can see, the socio-ecological resilience perspective already acknowledges the importance of people in creating resilient systems.

To conclude, the ecology literature added significantly to the resilience construct. They transferred the concept of resilience from materials to ecological systems and stressed the importance of the interaction between subsystems to system resilience.

2.1.3 Psychology

Research, experiences and common sense all recognize the importance of people in organizational success and see them as assets (Pfeffer and Veiga, 1999). Therefore, we also need to discuss and understand personal resilience. This section will elaborate more about personal characteristics and strategies that make people more resilient.

(14)

14

potentially traumatic experience (Bonanno, 2004). These people appear to be more resilient that others. Although independent, literature about human resilience emerged around the same time as research on ecological resilience, and was almost completely focused on young people. The emergence of the resilience concept in both streams of literature at the same time would indicate that conditions were ripe to take the concept to different fields (Masten & Obradovic,. 2008). Human resilience originated from the developmental theory, which examines people’s behavior across the life span and traces the origin, precursor signs and time sequence of a disorder. The area in psychology that relates to resilience is called developmental psychopathology. Developmental psychopathology addresses the developmental differences between people in their response to adversity and stress (Ponomarov and Holcomb, 2009). This concept has got a multidisciplinary character and ranges from genetic and biochemical factors to affective, cognitive and social realms. The concept assumes that there is continuity and discontinuity during the course of development. Furthermore it tries to highlight the stressors that characterize the turning points in the development, and stresses the necessity for awareness of personal fluctuations that characterize the lives of people who have a ‘‘at-risk’’ status. Another important construct of developmental psychopathology is the competence versus incompetence dimensions, which emphasizes that competent behavior in social and economic areas can alter the negative effects of severe stressors. As opposed to incompetent behavior, which makes people more vulnerable (Garmezy and Masten, 1986). According to a behavioral therapist, behavioral competence can be improved by systematic reinforcement of positive behavior, training, assisting in role modeling and enabling shifts in attributional processes.

Human resilience also arises from risk literature as these scholars (Anthony, 1974; Garmezy, 1974; Murphy & Moriarty, 1976; Rutter, 1979; Wemer & Smith, 1982 in Masten and Coastworth, 1998) realized that children flourished in the middle of adversity. They concluded that children could teach us how to promote competence, reduce risk and shift the process of development in a more positive direction. Furthermore they concluded that resilience can be identified by means of two factors (Masten and Coastworth, 1998). First, an individual has faced a significant treat characterized by a high-risk status or exposure to severe trauma or adversity. Second, level of adaption after this treat is high.

(15)

15

been remarkably consistent in what predicts resilience in children and the environment. Table 1 presents an overview of characteristics that identify resilient children and adolescents. It turns out that relationships, caring prosocial adults, and good intellectual functioning are the most important characteristics in determining if someone is resilient (Garmezy, 1985; Haggerty, Sherrod, Garmezy, & Rutter, 1994; Luthar & Zigler, 1991; Masten, Best, & Garmezy, 1990; Rutter, 1990; Wernet & Smith, 1982, 1992; Wright & Masten, 1997; Zimmerman & Arunkumar, 1994 in Masten and Coastworth, 1998). The characteristics mentioned in table 1 are not per se of causal influence, but are just associated to resilience. These characteristics could also be valuable traits of employees within an organization. We will elaborate more about these characteristics in other sections.

Table 1: Characteristics of Resilient Children and Adolescents Characteristics of Resilient Children and Adolescents

Source Characteristics

Individual Good intellectual functioning

Appealing sociable, easygoing disposition Self-efficacy, self-confidence, high self-esteem Talents

Faith

Family Close relationship to caring parent figure

Authoritative parenting: warmth, structure, high expectations

Socioeconomic advantages

Connections to extended supportive family networks

Extra familial context Bonds to prosocial adults outside the family Connections to prosocial organizations Attending effective schools

Source: Masten and Coastworth, (1998, p.207)

(16)

16

stress. Emotional self-awareness makes people more resilient because these people are able to detect stress before the stressors exercise their full impact, while people who have high levels of emotional expression can easily release stress. Emotional self-controlled people can cope with stress because they can control their impulses (Armstrong et al. 2011).

Above we discussed the characteristics of resilient individuals. Other scholars, such as Bonanno (2004), Luthans et al. (2006), Masten and Coastworth (1998) focused on how to develop resilience, which can be divided into reactive and proactive strategies. To successfully implement these strategies, it is important to understand the three core principals of human resilience: control, coherence and connectedness (3C’s). The 3C’s originated from social sciences and are built on the assumption that people have a natural capacity to recover and improve their adaptive capacities when they experience adverse conditions (Miller, 2005). First of all, people with higher levels of personal control in their lives are perceived to be more satisfied, less likely to have depressions, higher morale and they tend to get older. Control over stressors also improves people’s cognitive and motivational performance, lowers levels of stress reactivity, and improves a body’s ability to produce a normal immune response (Reich, 2006). Furthermore, coherence is fundamental to human resilience although it seems to be less apparent. People have the desire to remove uncertainty even in situation were information is not readily available, also called epistemic behavior. Finally, connectedness is important after disruption, because strong social ties show positive benefits such as better cardiac functioning, a prolonged life, less illness, and it can be an important moderator in buffering stress (Reich, 2006).

(17)

17

about the ability of a person to preserve a stable equilibrium in terms of psychological and physical functioning. Resilient people, in contrast to recovering people, encounter only a small change in their normal function (e.g. several weeks of restlessness or sporadic preoccupation) but generally display a stable trajectory after a disruption. Recovery includes the period of returning to the pre-event equilibrium which can take as long as one or two years (stability concept of Holling). Figure 2 shows and explains the difference between the two concepts.

Figure 2: Resilience vs. Recovery

Source: Bonanno (2004, p.102)

(18)

18

To conclude this section, Stewart et al. (1997) discovered, through an extensive literature overview, several common pillars of psychological resilience, which capture the essence of the concept perfectly:

 Resilience can be perceived as a dynamic process that is determined by life context  Resilience is a complex interaction between an individual’s characteristics and its

broader environment

 The level of resilience improves when there are less risk factors.  Resilience is important in life changes, and can be developed.

All these different features of personal resilience are also important for understanding organizational resilience. Organizations consist out of people and are often important in their ability to react to disruptions. Therefore, the people within organizations need to be resilient in order to create system-wide resilience.

2.1.4 Economic resilience

Organizations are part of a bigger picture, an economic system. The definition of economic resilience is constructed through a collaboration between social and natural scholars, which studied the sustainability of ecological-economic systems. Therefore, theories of economic resilience have many parallel concepts to evolutionary biology (Hanley, 1998). Economic resilience can be defined as ‘‘ the policy-induced ability of an economy to recover from, or adjust to the negative impacts of adverse exogenous shocks and to benefit from positive shocks (Briguglio et al., 2009, p. 233).’’

A resilient economic system creates socially correct outcomes on both local and national levels. To create a resilient economic system, inequalities (e.g. education, income and health) among the population should be redeemed and the growth of the core economy should be supported by institutional and macro level actions. The actions include building social capital and reducing vulnerability to environmental disruptions. Economic systems are closely linked with social and ecological systems since economic systems are meant to improve citizens’ well-being in a socially correct and environmentally sustainable way. Well-being encompasses a feeling of competence, autonomy, vitality, support from others and connectedness to other individuals.

(19)

19

different scales (local, national and global) according to Holling et al. (2002). Globalization even enhanced the interdependence and connectedness between economic and socio-ecological systems in such a way that actions of multinational organizations, institutions and governments could have a direct effect on the economic paths of regions. Decisions and policies taken in one place could have direct consequence for other places in the world. Diversity is a key concept in economic resilience, but globalization reduces it in terms of technology, biodiversity, language, tastes, values and preferences. Other characteristics of economic resilience are the ability of attracting FDI, skilled labor, competition, innovation, and flexibility in networks and growth. (Greenham and Ryan-Collins, 2013). Resilience phenomena do not always have a positive effect. Take the next example: the third world has high rates of fertility that are triggered, amongst others, by deficiencies in effective capital markets. Therefore, social security is not readily available for many households and having more children is the only solution. Consequently, increases in population density lead to higher risks of transmitting diseases, which make families more vulnerable to diseases. Then again, families respond by having even more children to avoid extinction. This example also shows that many systems are intertwined (Hanley, 1998).

2.2 Resilience in business

The concept of resilience can also be found in the business literature. Today, a weekday edition of The New York Times comprises more information than the average person in urban London during the 17th century encountered in a lifetime. This highlights the shift from a framework characterized by predictable, planned and evolutionary change to one of rapid and discontinuous change (Horne and Orr, 1997). The 21th century brought various challenges and changes for organizations and societies. Individuals have changed in all societies and are perceived to be different than previous generations: people that enter the job market today are different from those of 10 years ago. This generation is generally higher educated, more internationally oriented, completely familiar with technology and technological change, and they can better cope with large amounts of information. Moreover, more females are entering the job market.

(20)

20

quickly access global knowledge, are more collaborative, and have high expectations. Furthermore, they are more likely to accept diversity and have different interests and motivation. This generation, in contrast to the previous ones, sees changes increasingly as a positive thing, but is more able to deal with these changes.

This also means that leadership styles change. Take for example authority: positional power is not enough anymore to guide a company, because the new generation is better educated and less obedient. Therefore, the way of delegating, motivation, goal orientation, structuring and drivers has to change.

Furthermore, communication technology provides more opportunities, but also challenges for organizations in the 21th century. One of those challenges is information overload that could drown the recipients, and hampers organizational systems. Moreover, organizational development has become more dynamic; new sectors and organizations emerge and are forced to change more rapidly. All these changes require organizations and employees to evolve faster than ever before. Importantly, organizations detect change, anticipate the impact and use their competitive advantage to grasp the moment and start changing (Strycharczyk and Elvin, 2014). This highlights the amplified importance of organizational resilience in the 21th century.

(21)

21

Horne and Orr (1997) state that many factors of organizational resilience are embedded in their employees and processes, and that those factors are waiting on a supportive impulse to surface them. Organizations need people that respond effectively and quickly to change. At the same time their employees need to bear minimal stress during change, since these adaptive capabilities become more and more important in differentiating themselves from competition. When organizational resilience literature started to emerge, Mallak (1998) conducted a study to how individuals could improve resilience within organizations. He found seven principles that made people within organizations more resilient:

1. People should perceive experiences, even negative ones, constructively and find the positive perspective in it and move forward.

2. Responses should not be programmed but adaptive to the needs of the context.

3. Organizations should ensure access to suitable resources to enable a positive response to wide variety of events of its employees.

4. Organizations should expand the decision-making-authority to enhance adaptive responses and resource availability.

5. Develop capabilities to create on-the-spot solutions with immediately available materials.

6. Manager should encourage decision-making in situations where information levels are below the desired standard.

7. Individuals should be able to fill in whatever is needed in a team and have a common understanding of the team’s mission.

Lengnick-Hall et al. (2011) argue that capabilities, actions and interactions of individuals and units within organizations attribute significantly to the organizations capacity for resilience. Therefore, human resource (HR) policies, practices and activities provide the base of an organizations’ resilience capacity. According to Lengnick-Hall, there are three elements of organizational resilience that partly overlap the principles of Mallak (1998): cognitive elements, behavioral elements and contextual elements.

(22)

22

Behavioral elements consist out of resourcefulness, counterintuitive actions, practical habits and behavioral preparedness and lead to two key outcomes. First, a wide range of potential strategic actions can be used in various emerging situations, which are generated by the combination of resourcefulness and counterintuitive actions. Second, behavioral preparedness and practical habits produce routines that create options in people’s initial and instinctive response to various changes.

Interpersonal connections and resource availability are created by the following contextual elements: psychological safety, deep social capital, diffusion of responsibility and power and a broad resource network.

Strategic human resource management creates these competencies among the core employees of organizations. The strategic human resource system consists out of three components: HR principles, desired employee contribution, and HR policies (Figure 3). HR principles are guidelines that align less abstract policies and practices, while desired employee contributions consist out of attitudes, behaviors and work-related outcomes of employees that help them to contribute to implementing strategic objectives. These two components lead to HR policies, which reflect ways to complete specific employee objectives and guide the HR principles (Lengnick-Hall et al., 2011).

Figure 3: Strategic human resource management system

Source: Lengnick-Hall et al. (2011).

(23)

23

where trust is indispensable. Leaders play an important role in implementing this approach in the culture of the organization. They need to accept and believe that nobody, including himself or herself, are experts in responding to organizational challenges and events. Good leaders need to display, understand, openly discuss and develop trust within organizations.

Within the 21th century, leaders can and do not provide all answers anymore. Leaders need to entice the best out of their people and teams while being truly self-aware and understand situations for what they are (Strycharczyk and Elvin, 2014; Longstaff and Yang 2008; Lele, 1998). One party will not do business with another, unless they trust them to fulfil its responsibilities. ‘‘Trust is a mechanism for widening the basic unit of self-interest, and thus tightening diffuse feedback loops’’ (Lele, 1998, p. 231). This implies that systems cannot function properly or be resilient if there is no trust among the actors within it.

(24)

24

Complex systems, like organizations, are often capable of self-organization: dynamic and non-linear (Fiskel, 2003). The degree to which a system is sensitive to the effects of a disruption is called vulnerability. Gallopin (2006) argued that vulnerability consists out of three features: system sensitivity, exposure and the capacity of response to disturbances (Figure 4). His work stems for ecological literature but is also applicable on the concept of organizational resilience (Bhamra et al., 2011). Organizational resilience, together with adaptive capacity, is seen as a component of response capacity. Resilience is also a function of adaptive capacity. A systems capacity of response encompasses the ability to adjust and moderate the effects of a disturbance. While at the same time taking advantage of the opportunities it provides and cope with the consequence of the system changes.

Figure 4: The concept of vulnerability

Source: Bhamra et al. (2011, p. 5386)

(25)

25

different business strategies. Efficient organizations encourage decision and resource efficiency; while adaptability relates to learning abilities and cash reserves. Finally, cohesion encompasses a strong corporate culture and partnerships (Fiksel, 2003). Organizations can obtain different levels of resilience, just like ecosystems and human beings (Figure 5). The resistant system can handle a small disruption, but is unable to deal with high-impact disturbances. Furthermore, a resilient system can cope with wide range of disturbances and still perform their main function. In the end, the system will return to the stable state. Finally, the third system state represents the most resilient systems and these systems are able to cope with high-impact events. These kind of systems have multiple equilibrium states and can shift to other system structures and function (Bhamra et al., 2011).

Figure 5: Contiguous system states

Source: Fiksel (2003)

(26)

26

Figure 6: Resilient response framework

Source: Burnard and Bahmra (2011)

The above mentioned resilient response framework provides a deeper understanding of organizational behavior during disturbances. This framework can possibly link resilience to other concepts of business strategy such as competitive advantage. The earlier mentioned adaptive cycle provided the basis for this framework. The most important phase in this framework is the critical period, which consist out of detection and activation and are interconnected. This period is a vital stage in the resilient response framework, since these two elements provide a bridge between the threat and organizational response. It recognizes and interprets disruptions effectively, links the critical information to the right authority or system, and determines the ability of an organization to react positively. The critical period can be conceptualized as the adaptive capacity of an organization. Detection relates to having control over environmental feedback, and assessing the operational conditions. These two factors determine the adjustment of the system response to a disruption. This creates awareness of internal and external system challenges and context, also called embedded intelligence. Activation can only be effective if organizations can properly detect threats. During activation, an organization needs to efficiently mobilize capital and various resources (relational and cognitive) in times of environmental disturbance (Burnard and Bhamra, 2011). Eventually, if an organization is able to successfully pass the critical period, they will respond and adjust positively to a threat and eventually learn from it.

(27)

27 2.3 Resilience measurement

Measuring and benchmarking organizational resilience is important for the various reasons this thesis already discussed. It is useful for organizations because of two specific reasons. First, organizations can see how resilient they really are, and how they can improve their resilience. Second, organizations struggle to link resilience to disruptions and prioritize resilience in their organization. Measuring their level of resilience helps them to prioritize it (Stephenson et al., 2010). Lee et al. (2013) agrees to this theory and found that measuring resilience can help organizations to follow their resilience progress, to identify indicators of resilience, and see the link between improved resilience and competitiveness. Lastly, the organizations are able to recognize the need for resilience investments.

(28)

28

an extension of the tool mentioned before. They developed a tool with 15 indicators across three dimensions (Table 2).

Table 2: Resilience indicators Resilience indicator Situation Awareness Management of Keystone

Vulnerabilities

Adaptive Capacity

Roles and Responsibilities Planning Strategies Silo Mentality

Understanding of Hazards and Consequences

Participation in Exercises Communications and Relationships Connectivity Awareness Capability and Capacity of Internal

Resources

Strategic Vision and Outcome Expectancy

Insurance Awareness Capability and Capacity of

External Resources

Information and Knowledge

Recovery Priorities Organizational Connectivity Leadership Management and

Governance Structures

Source: McManus et al. (2007)

A widely used tool to measure these resilience indicators is a survey (e.g. Armstrong et al., 2011; Lee et al., 2013; Stephenson et al., 2010; Pal et al., 2014). Therefore, this thesis also opted to use a survey. There are several criteria for developing a resilience measurement tool (survey). The survey need to provide an depiction of organizational resilience at a certain point in time, and has to show the progress of an organization in getting more resilient. These criteria for a resilience measurement tool are the following (Stephenson et al., 2010; Windle et al., 2011):

 Parsimony: the tool should explain resilience in the simplest way possible. This could be done by a factor analysis, which simplifies the tool by excluding and combining questions. This will improve the accuracy of the tool.

 Validity: refers to the extent to which resilience is comprehensively measured by the items in the measurement tool. There are three kinds of validity: content criteria and construct validity. Content validity relates to the fact if all facets of the resilience concept are measured. Criterion validity tests are used to measure if the outcomes of the tool relate to a golden standard. Construct validity concerns the extent to which the tool outcomes are consistent with the hypothesis. The latter can be measured by performing a pilot study or expert peer reviews.

(29)

29  Usability: The tool should be easy to use. Difficulties can be caused by language

problems, access to the survey and technical issues.

 Applicability: The tool should be suited for different organizations in different sectors and regions.

When researchers develop a measurement tool, they should first define the indicators of organizational resilience by an extensive literature review like the ones in Table 2. Then, the components of these indicators should be identified. Finally, at least three questions should be developed per indicator and these should be reviewed by professionals and academics, and tested by a pilot (Stephenson et al., 2013).

This section elaborated about how to measure resilience and why organizations should track their resilience levels. The next section will formulate the hypothesis for this study

2.4 Hypothesis

This thesis will provide a five stepping-stone hypothesis. This method is derived from organizational performance, innovation and change literature but also the existing resilience literature up to now. Organizational resilience is a relatively new topic in business literature and no ‘‘grant theory’’ is developed. This thesis will provide a solid base for future studies and the subsequent development of a grant theory by investigating the influence of two organizational and two personal characteristics on organizational resilience.

2.4.1 Firm age

(30)

30

consequently be less resilient. On the other hand, research shows that organizational experience and experiences of other comparable organizations provide lessons and knowledge, which can be retained, notwithstanding time and turnover, through stable practices and routines (Argote et all,. 1990; Haunschild and Rhee, 2004; Haunschild and Sullivan, 2002; Levin, 2000; Levitt and March, 1988 in Desai, 2015). Experience can offer substantial benefits for organizations such as production cost and waste advantages, quality and reliability improvements, and enhanced success rates. Furthermore, the probability of organizational failure decreases by buffering activities, expertise for crisis management and accumulative knowledge (Desai, 2015). Also, efficiency of older organizations could be an asset in a stable environment but a liability in unstable situations (Thornhill and Amit, 2003). Moreover, older organizations are often better able to withstand disruptions due to their accumulated asset stocks (Amburgey et al., 1993). Another difference between younger and older firms can be found in the level of general management skills. As time passes on, the general knowledge of managers about matters like competitors, customers and suppliers becomes deeper and broader (Thornhill and Amit, 2003). Desai (2015) argues that experience could also hamper resilience in a way that if organizations face rapid and severe changes, they are more likely to engage in strategies that proved to be successful in the past, which in turn lower performance and survival. Nevertheless, this paper expects that older firms are more resilient that smaller firms.

Hypothesis 1: The age of a firm is positively related with the level of organizational resilience.

2.4.2 Firm size

(31)

31

(Buzzelli, 2001 in Garmestani et al., 2006). Large firms can more effectively capitalize on economies of scale than SMEs, and through their reputation effect they can maintain their status in the public consciousness and financial markets (Sleuwaegen and Goedhuys, 2002). Furthermore, although larger firms have slower growth rates, they are more prone to survive (Sutton, 1997). This could imply that they are more resilient than smaller firms. Other reasons why larger firms are more likely to survive is the radically increased competition among SMEs in the last few years, due to the technological advance and the globalized market and operations. Therefore, SMEs are forced to reassess some of their strategies and they are increasingly dependent on a variety of factors including resilience (Gunasekaran et al. 2011). Salavou et al. (2004) argued however, that SMEs have several advantages compared to larger firms: they are more able to adapt to change due to their size and flexibility. They are also more innovative and resilient because they face greater levels of competition in their markets. On the other hand, SME’s lack the resource for investments needed to compete with larger firms. That is why they are more likely to operate in niche markets that fit their capabilities to a fuller extend. However, this does not mean that SME cannot outperform larger firms. Their variable growth rates lead to mixed success in combating the larger firm (Garmestani et al., 2006). Nevertheless, this thesis expects that larger firms, in general, are more resilient that smaller firms.

Hypothesis 2: The size of a firm is positively related with the level of organizational resilience

2.4.3 Education manager

(32)

32

managers have high intellectual potential to learn and accumulate tacit knowledge. This would imply that they could better identify proper reactions to disturbance (Hitt et al., 2001). Education is valuable in the discovery and exploitation of various opportunities because it increases the stock of skills and information, including those mandatory to recognize and pursue opportunities successfully. Furthermore, high levels of managerial education can be related with greater receptivity and open-mindedness to innovation (Hambrick and Mason, 1984; Marvel and Lumpkin, 2007). These characteristics are also important when managers have to react to disturbances. Scholars (Bates, 1990; Bruderlet al., 1992; Gimeno et al., 1997 in Colombo and Grilli, 2005) found that the level of education is positively related to the likelihood of firms, especially new firms, to survive and reach high growth rates. Therefore, this thesis expects that the level education of a manager also is positively related to the level of organizational resilience.

Hypothesis 3: The level of education of a manager is positively related with the level of organizational resilience.

The level of education also seems to have more impact in larger firms than smaller firms. This is due to the fact that these organizations have often more operational complexity, and therefore need more knowledge and planning (Lee and Tsang, 2001). Von (2005) even argued that managers in small and medium enterprises (SME) have inadequate skill to effectively reflect upon their organizations. The Schumpeterian theory argues that organizations are more innovative if they expect high return on investment, and consequently monopolistic situations are an incentive to be innovative. This theory states that large firms have the resources needed, and possess a monopolistic power that enables them to be innovative and effectively use the manager’s knowledge base (Caloghirou et al., 2004). This could also apply to organizational resilience in a way that managers have more possibilities in their reaction to certain disturbances. Therefore, this thesis expects that firm size improves educational effectiveness of a manager.

(33)

33 2.4.4 Experience manager

Another element of human capital is experience. Experienced managers are perceived to have more self-esteem, feel more in control and are not afraid to fail. This is because they already have encountered change, and they were able to adapt to these circumstances and learned from their mistakes (Ayala and Manzano, 2014). This leads to lower transaction costs due to the fact that managers are more confident about circumstances surrounding the contracts, which in turn implies that experienced managers can reduce market transaction costs better than less experienced managers (McGee et al., 1995). In addition, Kor (2003) found that managers’ past experiences, on firm, team and industry level, contributes to the ability of a team to capture new growth opportunities. Hitt et al. (2001) argued that less experience managers are perceived to be less effective team leaders. Also, management experience will shape imagination, knowledge and improves the ability of a manager to assess employee competences. Past managerial industry knowledge of the threats, opportunities, competition and technology also contributes to firm growth due to the fact that developments are often path-dependent (Oster, 1999 in Kor, 2003). This obtained tacit knowledge helps managers to identify and assess changes in the industry environment and their response. Tacit knowledge could even result in a competitive advantage for the organization (Hitt et al., 2001). An organization also benefits more from a manager that has previous functional experience in the same industry because they understand what they do not know and what they could learn from cooperation with others. More experienced managers are perceived to be more knowledgeable about different aspects of the organizations competitive strategy. Furthermore, they are more able to identify potential risks that less experienced managers (McGee et al., 1995). Managers are required to effectively manage the available resources and to achieve economies of scope. However, managing diversity involves substantial managerial shrewdness which is often created by experience (Hitt et al., 2001). All these arguments let us believe that organizations with more experience managers are more resilient than others.

Hypothesis 5: The level of experience of a manager is positively related with the level of organizational resilience.

2.4.5 Expectation and conceptual model

(34)

34

variables to organizational resilience. In the past, many authors studied the concept of resilience. However, little research has empirically tested the construct of organizational resilience. Looking at previous organizational literature, this thesis assumes that there will be a positive relation between firm size, firm age, experience of the manager and the level of education and organizational resilience. Figure 7 shows a conceptual model of the proposed hypothesis and the control variables that could influence the level of organizational resilience.

Figure 7: Conceptual Model

(35)

35

3. Research method

The section discusses how this study is conducted. This chapter is divided into 5 sections. First, the setting, survey design and sample will explain where and how this thesis gathered its data. Also, the sample size is discussed. Second, the dependent, independent and control variables are operationalized. Third, the statistical model describes how this thesis will test the hypotheses. Finally, homoscedasticity, multicollinearity and normality will be discussed in the section about method assumptions

3.1 Setting

All organizations in the world face disruptions in one-way or another. These disturbances can differ in strength and occurrence and arise on different levels (individual, organizational or system level). The research was conducted in the Netherlands. This thesis applies data collected by means of a survey administered by the Faculty of Economics and Business (University of Groningen). The survey was part of ongoing research investigating the causes and consequences of firm performance in the North of the Netherlands. The survey was mailed to 4500 firms operating in different sectors. This thesis uses a subsample of the survey database and uses the information from firms operating in the sectors: agriculture, and industry and construction. The response rate for the respondents in these three sectors was 10.6 percent.

Among the advanced nation-states, the Netherlands is one of the most open economies of all. According to the ICCs open market index of 2015, the Netherlands is, after Luxembourg and Belgium, the most open economy in Europe and the fifth in the world1. The organizations included are both private as well as public companies (provincial or municipal organization). Therefore, the Netherlands offers a very suitable research context for what we try to do here. Our study offers an opportunity to test whether managerial and firm characteristics matter for organizational resilience and hence, for inclusive growth – a key aim of Dutch policies.

3.2 Survey design and Sample

We used a questionnaire that provides insight into managerial perceptions of resilience as well as organizational characteristics, context, performance and strategies. In line with

1

(36)

36

common convention in international business and strategy research, we employ a convenience sampling approach that is appropriate for studies that primarily aim for hypotheses testing, as is the case in our work. A mail survey was implemented, following well-documented response facilitation approaches (Dillman 2000; Fowler 2009; De Leeuw, Hox, and Dillman 2008). We decided in favor of a mail rather than a web survey, anticipating negative experience with other survey methods for business research in general (Dennis 2003; Harzing 2000; Shi and Fan 2009) and those in the Netherlands in particular (Deutskens, de Ruyter, Wetzels, and Oosterveld 2004; Berkenbosch, 2011). This experience indicates that many owners of small and medium-sized business are not web-enabled or are not willing to answer questionnaires via the Internet – an issue that was also confirmed in the pilot-testing phase of our questionnaire. The mail method aligns with the results from a meta-analysis reported by Shih and Fan (2009), who show that the response rates of traditional mail questionnaires are superior to email surveys, regardless of other survey characteristics such as the use of reminders or incentives.

We accounted for personalization, a short and easy to understand questionnaire asking only for relevant information, appropriate business-like (black and white) lay-out and print, a friendly presentation of the research, salience of the topic, feedback incentives, guaranteed anonymity of the results, university sponsorship, business-relevant support, stamped and university-addressed return envelopes, publicity of the research, pre-testing and timing of the survey, and an integral follow-up. The research was framed in terms of the expected demographic changes in the country, inducing a need for more entrepreneurship and public-private collaboration. It was presented as a joint effort of the local university, the chambers of commerce and the leading employer associations.

(37)

37

and questionnaire. The survey confirmed the accuracy of the database. Only 38 questionnaires were undeliverable, primarily due to a relocation of the company or bankruptcy (0.8 per cent). After the two waves, accounting for occasional cases with missing values or outliers, 475 respondents had replied, yielding a 10.6 percent response rate.

Of the respondents, 10.6 per cent confirmed that they are the owner or managing director of the addressed company. In total, 44.4 per cent of the respondents had an intermediate vocational education degree, and 32 per cent had a polytechnic or university degree. Taken together, this provides confidence that the respondents have sufficient knowledge and competencies to answer our questions. A comparison of responding to non-responding firms indicated no significant differences as to firm size and sector. We also compared early and late respondents. Armstrong and Overtone (1977) argue that late respondents are representative of non-respondents. We found no significant differences between early and late respondents on characteristics such as firm age, number of employees and work experience of the respondent

We used Harman’s (1967) single factor test to assess whether or not our data may feature significant common variance (Podsakoff and Organ, 1986). Unrotated factor analysis using the Eigenvalue-greater-than-one criterion revealed ten factors, with the first factor explaining only 10 per cent of the variance in the data. A principal component analysis resulted in seven factors. If a substantial amount of common-method variance were present, the factor analysis and the principal component analysis would have resulted in a single factor or one general factor accounting for the majority of the covariance among the variables (Podsakoff and Organ 1986). So, in our case, it is unlikely that the findings can be attributed to common-method bias.

3.3 Operationalizing variables

(38)

38

overcome interpretation biases in the survey. This thesis operationalizes the variables that are expected to be of direct consequences on organizational resilience.

3.3.1 Dependent variable

There is an ongoing debate on how to measure firm resilience, with a focus on the pros and cons of financial vis-à-vis non-financial and objective versus subjective indicators. With privately owned firms, precise financial measures are frequently unavailable. Due to their size and legal status, many companies are not required to publicly or otherwise report financial data. Therefore, organizational resilience studies increasingly rely on opinions of managers, following studies that revealed that the correlation between objective and subjective measures of performance tend to be high (Brouthers 2002; Dess and Robinson 1984). Moreover, in the business literature, it has been argued that enterprises form their strategy and competitive maps on perceived information and events, making subjective performance assessments by key decision-makers essential (Lang et al. 1997). Both arguments imply that subjective perceptions are valid performance measures, being reliable and having material consequences. The survey measures organizational resilience with 4 questions. These questions are:

1. Our organization is very dynamic and entrepreneurial 2. Our employees are willing to take risks

3. Our organization emphasizes revenue growth by new product introductions and investments

4. Open-mindedness toward new challenges is the main motto of our organization

Respondents could refuse to answer questions or might not be able to answer. In case of non-response, this thesis excludes the survey from the estimation. A five point-Likert scale is used to answer the various questions of resilience, which ranges from strongly agree (1) to strongly disagree (5). To create one measure for organizational resilience, a factor analysis was performed to show the uni-dimensionality of the four different items. We also calculated the Cronbach’s alpha of the different items to assess how much the different indicators measure the same underlying construct (Cronbach’s alpha: 0.78, appendix 2, table 1). Both values indicated that we could create one index of organizational resilience by combining the four items.

3.3.2 Independent variable

(39)

39

the dependent variable organizational resilience. The independent variables are firm size, firm age, experience of manager and education of manager.

Firm age (H1): Firm age is measured as the year of foundation. The question asked to respondent is ‘‘in which year is your business or business unit established?’’

Firm size (H2, H4): The current amount of FTEs (fulltime equivalent) in the organization. Respondents were asked ‘‘how many full time employee (FTEs) are currently working in your business?’’

Education manager (H3, H4): (1) lower or elementary education, (2) intermediate vocational education degree, (3) polytechnic degree and (4) university degree. Managers were asked ‘‘what is the highest form of education you completed?’’

Experience manager (H5): Total paid and professional work experience of manager in years. The question asked to measure the experience of the manager is: ‘‘in total, how many years of work experience do you have (Paid and professional)?’’

3.3.3 Control variable

The validity is tested by several control variables that are kept constant. The control variables used in this thesis are organizational performance, revenue, multiple sector engagement and municipal quality.

(40)

40

The second control variable is revenue. Revenue is measured by the average revenue in the last two year and could influence organizational resilience. Organizations that have substantially higher revenue than others often have more resources available to withstand disruptions. Revenue is measured through an interval scale: (1) less than 100.000, (2) 100.001 – 250.000, (3) 500.001 – 750.000, (4) 50.001 – 1.000.000, (5) 1.000.001 – 1.500.000, (6) 1.500.001 – 2.500.000 and (7) more than 2.500.000.

The third control variable is related to the quality of the municipality an organization is operating in. We asked respondents to rate their municipality on a five point Likert-scale ranging from 1= totally disagree to 5= totally agree on 6 characteristics:

1. The municipality makes quick definitive decisions. 2. The municipality is solution focused.

3. Officials in the municipality are easy to reach. 4. Decision procedures are flexible.

5. Municipal decisions are predictable.

6. The supplied information of the municipality corresponds with our needs

Differences and similarities of environmental conditions could be assessed through comparing environmental performance of the different municipalities (Ferrarini et al., 2001). Environmental conditions have a substantial influence on organizations and could facilitate improved resilience. Therefore, this thesis uses municipal quality as a control variable. We constructed one variable for municipal quality. The factor analysis showed good results and the Cronbach’s alpha was 0.77 (appendix 2, table 1). However, if we deleted question 5 (the decisions of the municipality are predictable?) the Cronbach’s alfa showed improved results and increased to 0.82. Therefore, we excluded this measure from the aggregated variable.

(41)

41

Table 3: Variables, definition and measure

Type Variable Definition Measure

Dependent Independent Independent Independent Independent Control Control Control Dummy/Control Organizational resilience Firm age Firm size Education manager Experience manager Organizational performance Revenue Municipal quality Multiple sectors

Organizations adaptive capacity during turbulent periods.

How many years the firms exists

Amount of FTEs in currently in the organization

Level of education

Years of work experience

Relative performance in market share, profit, employment and investments

Average revenue in the last two years

Overall quality of the municipality in which the organization is operating in.

Company is active in multiple sectors

Five point Likert-scale: (1) strongly disagree, (2) disagree (3) neutral (4) agree (5) strongly agree

Year of establishment

Amount of FTEs

(1) lower education, (2) high school education (Mavo/vmbo/havo/vwo) (3) secondary professional education (MBO), (4) higher professional education (HBO) (5) university degree.

Years of work experience (paid and professional)

(1) Less than 100.000, (2) 100.001 – 250.000, (3) 500.001 – 750.000, (4) 50.001 – 1.000.000, (5) 1.000.001 – 1.500.000, (6) 1.500.001 – 2.500.000 (7) more than 2.500.000.

(0) One sector (1) Multiple sectors

3.4 Statistical model

To test our hypothesis, we us the following statistical model:

Organizational Resilience (y) = β0 + β1X1 + β2X2 + β3X3 + β4X4 + βcXc + ε

Why are some firms more resilient than others? The effect of firm and managerial characteristics on organizational resilience