Repetitive reorganizations, uncertainty and change fatigue

ABSTRACT

Repetitive reorganizations cause change fatigue, which creates employee resistance to further organizational change. This paper investigates how such fatigue arises. The results show that uncertainty and workload are mediating factors. The effect of change fatigue is not moderated by the perceived success of the prior reorganization, participation in that process, or leadership characteristics. It is only slightly moderated by the satisfaction of employees about the communication during the prior organizational reorganization.

IMPACT

Resistance against organizational reforms is commonly misunderstood. Whereas the academic literature suggests minimizing resistance by coercive means, eventually crushing it, this paper suggests something entirely different. Change fatigue as a form of resistance to upcoming reorganizations emerges from experiences with earlier reforms. Each time a new major organizational change is planned, change fatigue among personnel increases. Therefore, drastic changes should only be a last resort where no alternative solution can be employed. If reorganizations are necessary, the prime concern should be to take measures to avoid and reduce uncertainty among employees, preferably by good communication.

KEYWORDS:

• Change fatigue
• reorganizations
• transformative leadership
• uncertainty
• workload

Introduction

Reorganizations can lead to employee resistance, and overcoming such resistance is a major topic in management studies (Lewin, 1947; Daft, 2014, p. 365; Price, 2019). The motives for such resistance are primarily perceived to be political-strategic, materialistic, and/or economic.

Political-strategic resistance to reforms occurs when people do not share the vision, goals, or instruments that the upcoming reform promises, or when the reform creates changing power relations, conflicts of interests, and/or conflicting values (Lewin, 1947). Materialistic resistance arises when employees fear that the reform will undermine their personal interests; that increasing job demands and decreasing resources will render them unable to meet expectations; that they will lose control over their situation; that they will lose their job; and/or that their working conditions will deteriorate (Dent & Goldberg, 1999). Economics-based resistance results when employees weigh the costs and benefits of reform and perceive a negative outcome. In this case, employees judge the reform to be futile, or even detrimental, to the organization (Dubrin & Ireland, 1993).

Each of these theories assumes that such resistance is focused on the nature and impact of the upcoming reform. The mainstream academic literature holds that ‘The causes of resistance are primarily attributed to the shortcomings of employees, such as self-interest, a lack of tolerance for change, cynicism or fear i.e. inappropriate attitudes and behaviors on the part of employees gives rise to resistance which, in turn, impedes the change initiative’ (Ybema et al., 2016, p. 7). The answer to such resistance can be summarized in the statement that ‘It is a problem either to be eliminated or minimized’ (Giangreco & Peccei, 2005, p. 1816). If employees fail to respond in a satisfactory manner (for example by ceasing to resist), managers are justified in using more coercive methods to impose the change. The aim is to stamp out resistance and ‘correct’ those intransigent individuals who engage in it (Ybema et al., 2016).

This paper disputes this assumption, as well as its proposed solution. It argues that the literature neglects ‘change fatigue’. As a result of repeated reorganizations, employees become alienated from the organization (Weber & Weber, 2001; Pollitt & Bouckaert, 2014). They might even reach a point of exhaustion, resulting in an incapacity to support and adapt to further change (Bernerth et al., 2011). The existing literature on the topic does recognize change fatigue as a source of resistance (O’Toole, 1995), but research into resistance has mostly ignored the unintended consequences of previous reforms (Wynen et al., 2019, p. 699).

This paper addresses the research questions ‘Do repetitive organizational reforms result in change fatigue, how is this mediated, and what can be done to moderate this effect?’

We answer these questions by first examining the dominant theories on resistance to organizational change. We also explore an alternative view, which proposes that the frequency of previous reforms is of central importance. The sub-question is ‘How do mainstream theories explain resistance to organizational change, and how does this differ from a theory in which previous organizational changes are seen as determinative for such resistance?’

We then present an empirical analysis of a European financial institution—we will call it ‘Bank 2.0’, as it wishes to remain anonymous. Previously a commercial institution, Bank 2.0 became nationalized at the onset of the 2007 financial crisis and, after numerous reforms, its stock is now traded on the private market. The first author of this paper was permitted to make an assessment of this organization from the inside, under the condition of confidentiality, for six months. During this time, she surveyed the company’s personnel. Multivariate analysis of the data points to a strong relationship between the number of reorganizations experienced by employees and the degree of change fatigue. An extended analysis shows that this relationship is almost inevitable.

Finally, we address the implications of this research. If resistance to organizational change is due to experiences with previous organizational changes, what does this imply for handling such resistance?

Theory

Organizational changes can have unintended consequences. Researchers have pointed to emotional exhaustion, stress due to uncertainty, increased turnover rates (Bernerth et al., 2011; De Vries, 2013), and absenteeism (Wynen et al., 2019). This side effect of reorganizations can be called ‘reorganization fatigue’, ‘reform fatigue’, or ‘change fatigue’. The last term is the most neutral and, following Bernerth et al. (2011) and Dilkes et al. (2014), we will use that term here. Bernerth et al. (2011) define change fatigue as ‘a perception that too much change is taking place’ (p.322). Employees suffering from change fatigue show inclinations and behaviors that suggest they do not embrace the upcoming change. They also demonstrate an attitude counter to the reform’s goals and might even revolt against the reform. Change fatigue is indicative of psychological motivations to resist organizational reforms. It reveals a preference for stability; organizational changes involve uncertainty, which people tend to avoid. According to Lora et al. (2003), indicators of change fatigue include a lack of support for new changes, loss of confidence in leadership, and a less proactive attitude towards further changes (p. 4).

Price (2019) offers a more formal explanation, arguing that a ‘change equation’ identifies three key components of an individual’s reaction to change. These three components are the degree to which an individual is satisfied or dissatisfied with the current state; the degree to which the change is desirable; and whether or not change is practical. According to Price, ‘Individuals relate the three elements directly to the perceived cost to self … individuals will determine their degree of resistance in relation to the importance they attach to those aspects of the required change that disrupt their satisfaction with the current state’ (p. 16). Change fatigue, as one of many possible forms of resistance, emerges when employees oppose any further change, regardless of their satisfaction with the current state and their opinions about the goals of the change.

Change fatigue is the consequence of multiple organizational reforms (Vestal, 2013; Dilkes et al., 2014; Frahm & Brown, 2007; Bernerth et al., 2011). It predicts resistance to upcoming reforms (De Vries, 2013) and is characterized by exhaustion and an inability to adapt, implying that the employees become tired of innovations and reform initiatives and the way these reforms are implemented (Garside, 2004; Kouchaki & Desai, 2014).

Reasons therefore exist to suggest that change fatigue is a real phenomenon, but what does an explanatory theory look like? Before describing this theory, we must first discuss organizational reforms.

Weick and Quinn (1999) see reorganizations as radical, even revolutionary, changes. Reorganizations have also been called ‘planned episodic changes’ (Burke, 2017; Weick & Quinn, 1999). Such revolutionary change is most often planned from a top-down perspective. Predefined goals guide the process, while employee participation follows and is mostly advisory in nature (Burke, 2017).

Reorganizations differ from process changes. The latter are more incremental, similar to evolutionary changes (Weick & Quinn, 1999). Process changes are small, often unplanned, and continuous, with goals that only become clear afterwards from discussions on the shop floor, thus reflecting a bottom-up process (Burke, 2017; Weick & Quinn, 1999). In process changes, business processes are primary, and the position and role of employees is often secondary. The impact of such processes is small, as it often only involves working with a new information system or making changes to process-oriented roles or teams. This type of change appears to be more successful than reorganizations (Burke, 2017). Revolutionary change occurs quickly and affects virtually all parts of the organization, whereas evolutionary change occurs slowly and gradually and may involve only one area of the organization (Malhotra & Hinings, 2015).

The characteristics of reorganizations (for example their top-down nature, their impact on jobs and personnel), in contrast, are why employees who have experienced many such reforms become anxious about a new reform.

Pollitt and Bouckaert (2011) see organizational reforms as purposeful changes to an organization’s structure and processes, aiming to improve the performance of the whole organization. Such reforms have drastic consequences for employees. For example, they might be told to change the way they work, re-apply for their job, be transferred to another position, or even be made redundant. According to Sistare (2004), the goals of such organizational reforms concern saving money; concentrating power; addressing urgent problems; and, in the public sector, strengthening and creating a better society through more efficient and effective service delivery, increased transparency, and improved employee performance. Reforms are said to be a jarring, but also a ubiquitous phenomenon: on the one hand, they are bemoaned for the fact that they often fail to meet the ambitious goals they were meant to address and are consequently ineffective, while, on the other hand, new reforms are steadily being initiated, as they are perceived to be successful as an instrument for planned change (Kleidat, 2011, p. 17).

Mediators

In theory, two mediating factors are responsible for this causal relation: workload and uncertainty. The job-related aspects of organizational reforms result in both increased workload and uncertainty among employees about how the upcoming reforms will affect them. Workload and uncertainty can both be expected to determine the degree of change fatigue.

We first expect that reforms increase employee workload, leading to fatigue. Reorganizations take time and effort from all involved. The regular day-to-day work must still be done, and the reform process comes on top of that. Work may occur under changing conditions, with new routines, new rules for accountability, higher demands on performance, a new work role, and new hierarchical relations in the organization. Employees may become overworked and therefore tired as they try to adapt to the change in addition to doing their regular work. This effect is recognized in the literature: ‘Staff, within organizations that are swamped by repetitive structural reforms, suffer from high levels of stress and anxiety’ (Spicer, 2018, p. 1). Research shows that repetitive reforms result in increased absenteeism, especially when the reforms come in quick succession. In that case, employees have no recuperation time between previous structural reforms and new ones (Wynen et al., 2019, p. 699). When a reorganization involves downsizing, workload increases, as the remaining employees must often do the same total amount of work with fewer people. The existing literature suggests that repeated reforms lead to a higher resignation rate, reduced employee cooperation, reduced morale, increased work stress, greater levels of disobedience, the initiation of employee lawsuits, anti-social behavior, and a growing distrust in the company (Appelbaum et al., 1999).

Scholars have also noted the unintended consequence of organizational reforms on employee uncertainty. Bernerth et al. (2011) see psychological uncertainty as a negative side effect of organizational reforms, because such reforms affect the capacity of individual employees to accurately predict the nature of their future job, their position in the organization, and what will be expected from them. Organizational reforms produce uncertainty in different ways (Ashford, 1988). Employees will become uncertain if they lack knowledge about the details of the reform (Burke, 2017). This causes anxiety about how the reform will affect them personally. Frahm and Brown (2007) add that it is awkward to be in a position of uncertainty, especially when one is accustomed to certainty in a job. When job security is at stake and it becomes evident that the nature of the job will change, employees become uncertain. Too many sequential or simultaneous changes can result in an inability to adapt, leading employees to feel uncertain whether they can meet future job demands. This uncertainty, in turn, leads to exhaustion (Bernerth et al., 2011).

Moderators

If we are correct in suggesting that the characteristics of organizational reforms, as opposed to the characteristics of incremental process changes, are responsible for the emergence of change fatigue, we might expect that if earlier reforms are successful, this effect can be avoided in the future. We can therefore expect that the probability that employees will experience change fatigue will be influenced by how satisfied they are with the outcomes of previous reforms (Bernerth et al., 2011). Employees have varying ideas and expectations about upcoming reforms in relation to their experiences with previous reforms. They project prior experiences onto their expectations for the upcoming reform. Lora et al. (2003) suggest that change fatigue at its core results from a failure of previous reforms to have met their goals, resulting in an almost automatic employee projection of this failure onto the upcoming reform.

The perceived success of prior organizational reforms can therefore be expected to affect the way employees perceive the likely success of a new reform (Atkinson, 2005; Hannan & Freeman, 1984). Since approximately 70% of organizational reforms do not achieve their goals, there is considerable cause for concern (Burke, 2017; Beer & Nohria, 2000). Pollitt and Bouckaert (2004, 2011) conclude that public sector reforms almost never achieve their goals, for example.

De Vries (2013) mentions that reorganizations are successful if they result in added value for the organization, its employees, and clientele. If that is not the case, and the costs exceed the benefits, change fatigue is likely. If the previous reorganization has been successful, it is less likely to have resulted in change fatigue than if it has not succeeded. The success of the previous reform can be seen as a moderator in the causal relation between the number of previous reforms, their intensity, and the occurrence of change fatigue.

A second moderating factor can be found in employee confidence in supervisors’ transformative leadership capabilities. Scholars have argued that transformative leadership is crucial in organizational reforms (Podsakoff et al., 1990; Vestal, 2013; Garside, 2004; Higgs & Rowland, 2011). They claim that it is leadership’s role to make changes in collaboration with employees rather than imposing changes top-down (Higgs & Rowland, 2011). According to Vestal, leadership must ensure that employees are involved in the change process, find solutions to employees’ concerns, and strive for employee support (Vestal, 2013). This kind of leadership is essential to successfully creating change, as it helps employees realize the need for change and the values that drive the change initiative (Podsakoff et al., 1990); inspires employees by identifying and articulating a vision; models exemplary behavior; supports employees; and offers intellectual stimulation (Podsakoff et al., 1990). At the very least, leaders must be positive about the upcoming change, communicate with substantiated arguments why the change is necessary, and involve all stakeholders (Dilkes et al., 2014). if such transformative leadership were visible during the prior reform, it would be expected to have a moderating effect on the causal relation between the number of reforms and change fatigue.

Employees must be informed about the reasons for the reform and the expected benefits (MacIntosh et al., 2010). This communication should reduce uncertainty (Frahm & Brown, 2007), as it helps employees understand why the reform is necessary and what to expect. If the reform’s justification, goals, and consequences are promptly and openly communicated, employees feel respected and involved and are more inclined to accept the changes (Weber & Weber, 2001; Frahm & Brown, 2007). This suggests that good communication serves as a moderating effect for change fatigue.

Participation in reform processes serves as a similar moderating factor. Coch and French (1948) concluded that leadership that lets employees participate in the reform’s design and development results in lower resistance. They advised managers to hold group meetings to communicate the need for change and to encourage employee involvement in planning the change (Dent & Goldberg, 1999). According to Pasmore and Fagans (1992), the key to establishing excellence and motivation among personnel is to make them participate in reform processes. Compensating the top-down perspective in reform processes by allowing bottom-up initiatives may increase the probability of success, as this process results in better solutions to existing issues and presents a self-enforcing cycle of commitment, coordination, and competence (Beer et al., 1990). Recent studies have also underlined the importance of participation during change processes (Burnes, 2015). These studies show that leadership can reduce resistance to change through intervention that increases employee participation in the process. For example, Scandura (2019) concludes, ‘People are more likely to accept changes that they help design’ (p. 736), and Ybema et al. (2016) state that ‘being kind to others makes it more difficult for them to resist, subtly pushing them into willing compliance’ (p. 12).

Looking at the literature this way, we came up with four hypotheses:

H1: Change fatigue is a consequence of the number of reorganizations experienced and their perceived intensity.

H2: The effects mentioned in H1 are mediated by the uncertainty and workload involved in reorganizations.

H3: The effects mentioned in H1 and H2 are moderated by the perceived success of the prior reorganization, the level of satisfaction about communication and participation in that process, and the level of satisfaction about existing transformative leadership in the organization.

H4: Including mediating and moderating effects in H2 and H3 adds to the explained variance in change fatigue.

Research method

To test our model and hypotheses, we conducted a web survey in 2018. The survey was sent to all team leaders in the business branch of Bank 2.0. That branch consists of six departments with a total of approximately 3,000 employees, of whom 340 responded. Because change fatigue was perceived as a sensitive subject and leadership feared that the survey would send the wrong signal since a reorganization was planned, team leaders were asked to approach employees and request their cooperation. This resulted in a response rate of just over 11%.

Of these respondents, 65% were male. Their average age was 44 years; they had been working at Bank 2.0 for 20 years on average, with 4.4 years in their current position. Twenty per cent of the respondents had a university degree, 50% had completed a polytechnic education, and 30% had only a high school education. On average, they had experienced seven reorganizations during their working life. On the items indicative of change fatigue (explained below), measured on a seven-point Likert scale, their average score was 4.98. The standard deviation was 1.16, indicating a serious degree of change fatigue on average. None of the respondents held a managerial position within the organization.

The survey consisted of single questions and item batteries, which were mostly replications of item batteries found in international journals. The scales had therefore already been validated in earlier research. However, we could not find a preexisting scale for the variable on the perceived success of prior reforms, so we had to develop the items for this factor ourselves. Table 1 provides the exact wording of the survey questions, as well as the reliability of the scales as measured by Cronbach’s alpha.

Table 1. Factor loadings of items used in the analysis.

Reform fatigue: Cronbach’s alpha = 0.856
1. Too many change initiatives have been introduced at Bank 2.0	.804
2. I am tired of all the changes in this company	.710
3. The amount of change that takes place at Bank 2.0 is overwhelming.	.858
4. We are asked to change too many things at Bank 2.0	.749
Uncertainty: Cronbach’s alpha = 0.901
1. I am often uncertain about how to respond to change.	.771
2. I am often unsure about the effect of change on my work unit	.823
3. I am often unsure how severely a change will affect my work unit	.917
4. I am often unsure about the future of my job	.737
5. I am often unsure about my job	.732
Workload: Cronbach’s alpha = .815
1. During the last years, I increasingly have to meet scheduled deadlines	.741
2. I have fewer and fewer time to finish the job	.728
3. I work more and more under time pressure	.853
Success of prior reorganization: Cronbach’s alpha = 0.885
1. The prior reorganization was implemented in a meticulous way	.805
2. The prior reorganization was effective	.628
3. The prior reorganization resulted in cost savings	.565
Confidence in transformative leadership: Cronbach’s alpha = 0.950
My supervisor …
1. … Inspires others with their plans for the future	.829
2. … Is able to get others committed to their dream	.862
3. … Provides a good model for me to follow	.826
4. … Leads by example	.836
5. … Challenges me to think about old problems in new ways	.807
6. … Asks questions that prompt me to think	.906
7. … Has stimulated me to rethink the way I do things.	.852
8. … Has ideas that have challenged me to re-examine some of my basic assumptions about my work	.798
Communication during prior reorganization: Cronbach’s alpha = 0.904
1. There was good communication between project leaders and staff members about the organization’s policy toward changes	.843
2. The corporate management team kept all departments informed about its decisions	.842
3. Two-way communication between the corporate management team and the departments was very good	.884
4. The corporate management team clearly explained the necessity of the change	.772
Participation during prior reorganization: Cronbach’s alpha = 0.876
1. Changes were always discussed with all the people concerned	.773
2. Decisions concerning work were taken in consultation with the staff who were affected	.726
3. Departments were sufficiently consulted about the change	.866
4. Staff members were consulted about the reasons for change	.805
5. Our department provided sufficient time for consultation	.718
6. Staff members were sufficiently involved in the implementation of the changes	.830

For the dependent variable, the items measuring change fatigue were copied from Bernerth et al. (2011); there were six items with response categories on a seven-point Likert scale (‘completely disagree’ to ‘completely agree’). The mediating factors included five items on uncertainty copied from Bernerth et al. (2011). The nine items on workload were copied from Eurofound (2015).

The two independent variables—the number and perceived intensity of experienced reforms—were measured by single questions. The first question was ‘During your total career, how many reorganizations have you experienced?’ This question was preceded by a definition of reorganizations. Secondly, we asked about the perceived intensity of the experienced reforms through the question ‘To what extent did those reorganizations fundamentally alter your work?’, which the respondents answered on a seven-point Likert scale.

To measure satisfaction with the transformative leadership of a respondent’s supervisor—one of the two moderating factors—item batteries were copied from the questionnaire used by Podsakoff et al. (1990) to measure transformative leadership.

The items for communication and participation during the most recent reorganization (see Table 1)—also on seven-point Likert scales—were copied from Bouckenooghe et al. (2010).

We developed six items to measure the perceived success of the prior reform, as no acceptable item battery was found in the literature. These items related to the speed with which the prior reform had been carried out, its effectiveness, the cost savings achieved, the resulting client satisfaction, and the innovation the reform delivered. These six issues were framed as statements with which respondents could agree or disagree on a seven-point Likert scale.

Bank 2.0

The banking system is highly competitive and dynamic, requiring continuous adaptations to changing contextual circumstances. This system therefore provides the context for a good case study. Bank 2.0 had a particularly strong need to adapt, because it was heavily impacted by the 2007 financial crisis and needed government support to survive. The bank, which had merged with another bank a couple years earlier, was nationalized and forced to split into two separate holdings.

Bank 2.0’s nationalization resulted in political demands for measures to prevent the same problems from happening again. The implementation of these measures created oversight divisions; adjusted the balance between diversification and concentration of activities in favor of concentration (by restricting acquisitions, and changing views and demands on investments in profitable politically neutral stocks, such as in oil or tobacco companies); and required the bank to retreat from the global playing field and concentrate on the European market. Most importantly, the bank needed to be restructured so it could rejoin the stock market in the near future. This was achieved by requiring savings, high returns on investments, and profit generation. These initiatives succeeded somewhat, but Bank 2.0 is still a partially nationalized financial institution.

These changes to Bank 2.0 came on top of an already highly dynamic banking environment, in which international organizations responded to ICT developments by implementing e-banking, interest rebates, increasingly strict regulations concerning anti-laundering policies, ‘know-your-customer’ policies, and requirements on liquidity ratios. (Basel III is an example of these measures.)

These dynamic contextual developments, as well as changes to the bank’s internal governance, led to frequent conversations about how to balance the need for bureaucratic controls with the need for flexibility to respond to market dynamics. Multiple oversight divisions were created, but these soon disappeared or had to be recombined. Bank 2.0’s various divisions experienced multiple reorganizations, sometimes within the same year, making the bank an appropriate case study for our analysis.

Analysis

This section presents the outcomes of our analysis, testing the theoretical framework presented above. First, we analysed whether the item batteries adequately measured the latent variables. This was done by confirmatory factor analysis using R. Three models were developed. The first model regressed change fatigue on the number of reforms experienced and their perceived intensity. The second model used two mediating factors: uncertainty and workload. The third model used two moderating factors: perceived success of the prior reform and transformative leadership. We analysed whether the effects on change fatigue were significant and what their predicted direction was, as well as whether these effects changed if mediating and moderating factors were included. We also wanted to know whether the total variance in change fatigue increased when the mediating and moderating factors were included, or whether these factors simply mediated and moderated effects already present.

Confirmatory factor analysis

The first part of the analysis optimized the measurement model regarding the five latent variables (change fatigue, uncertainty, workload, success of the prior reorganization, and satisfaction about transformative leadership). To do this, we first deleted all cases with missing values on one of the items used in the model. This reduced the number of respondents to 190 but resulted in a complete data file without missing values.

We next conducted a confirmatory factor analysis in R to estimate the parameters (for example factor loadings of items and the co-variances between the latent variables). The goal was to establish a measurement model that satisfied the statistical model-fit criteria proposed in the literature (for example χ-2, CFI and RMSEA and PClose) (Schmitt, 2011). This required deleting some measured items from the analysis and resulted in the measurement of change fatigue through four items, workload through three items, uncertainty through five items, satisfaction about transformative leadership through eight items, success of prior reorganization through five items, and communication during the prior reorganization through four items (see Table 1). This measurement model, with the factors eventually showing up in the final model, satisfied the fit criteria: χ-2 was 507.880 with 298 df and a p value equal to 0.10, which is above the threshold of 0.05; CFI was equal to 0.973, which is also above the threshold. The RMSEA was equal to 0.045, below the required threshold of 0.08 suggested in the literature, and the PClose was 0.671 above the required threshold of 0.5. The measurement model was therefore a good fit. The variance and co-variance matrix are shown in Table 3.

Table 3. Correlation matrix.

	Reform fatigue	Intensity of previous reorgan-izations	Number of previous reorganizations	Workload	Uncert-ainty	Leadership	Success	Communication	Partici-pation	Interaction between number of previous reorganizations and commun-ication	Interaction between number of previous reorganizations and participation	Interaction between number of previous reorganizations and leadership	Interaction between number of previous reorganizations and success
Reform fatigue	1
Intensity of previous reorganizations	,242**	1
Number of previous reorganizations	,335**	0,088	1
Workload	,216**	,248**	0,121	1
Uncertainty	,462**	,188*	0,073	,230**	1
Leadership	−,248**	0,023	−,213**	−0,071	-0,14	1
Success	−,201**	−0,128	−0,132	−0,029	-0,113	,339**	1
Communication	−,189**	−0,017	−0,078	−0,026	-,144*	,612**	,618**	1
Participation	−,160*	−0,073	−,253**	−0,1	-0,074	,563**	,504**	,622**	1
Interaction between number of previous reorganizations and communication	0,061	−0,024	−,231**	0,072	-0,041	0,032	0,013	0,057	0,013	1
Interaction between number of previous reorganizations and participation	−0,059	−0,059	−,294**	−0,025	-0,01	0,069	0,063	0,012	0,12	,509**	1
Interaction between number of previous reorganizations and leadership	−0,015	−0,052	−,257**	−0,034	-0,033	0,112	0,04	0,031	0,068	,636**	,597**	1
Interaction between number of previous reorganizations and success	−0,036	0,026	−,320**	−0,01	0,03	0,043	0,026	0,014	0,068	,613**	,550**	,462**	1

Estimating the effect parameters through structural equation modeling

This section presents the test of what was described in the theory section. Testing was done in three steps. The first model estimated only the direct effects of the number of reorganizations experienced and their perceived intensity on change fatigue. The two mediating factors—workload and uncertainty—were added in the second model. The moderating factors—prior reorganization success and satisfaction with the process—were added in the third model. Table 2 presents the standardized effects on change fatigue in these three models, including the mediating and moderating effects in the latter two models.

Table 2. Modeling the causal effects on change fatigue.

	Model 1: Without mediating and moderating factors	Model 2: Including mediating factors	Model 3: Including mediating and moderating factors
Total explained variance in change fatigue	0.19	0.40	0.43
Standardized direct effects on change fatigue	Effect (p value)	Effect (p value)	Effect (p value)*
No. of previous reorganizations	0.35 (.000)	0.33 (.000)	0.36 (.000)
Intensity of previous reorganizations	0.23 (.000)	0.09 (.206)	0.07 (.301)
Workload		0.07 (.353)	0.05 (.463)
Uncertainty		0.45 (.000)	0.45 (.000)
Communication during prior reorganization			−.012 (.066)
Interaction communication—No.of reorganizations			0.17 (.010)
Standardized direct effects on workload
No. of previous reorganizations		0.11 (.153)	0.12 (.153)
Intensity of previous reorganizations		0.17 (.035)	0.17 (.036)
Communication during prior reorganization
Interaction communication-no-of-reorg
Standardized direct effects on uncertainty
No. of previous reorganizations		0.13 (.719)	0.12 (.858)
Intensity of previous reorganizations		0.31 (.000)	0.30 (.000)
Communication during prior reorganization			−0.07 (.315)
Interaction communication-no-of-reorg			0.03 (.711)
Model fit
χ−2	8.558	98.999	177.136
Df	8	68	136
P	0.381	0.08	0.10
CFI	0.999	0.975	0.975
RMSEA	0.019	0.050	0.041
PClose	0.680	0.495	0.816
N	187	187	187

*p values are in parentheses.

The first model demonstrated strong and significant effects on change fatigue resulting from the number of reorganizations experienced and their perceived intensity, with standardized βs equal to 0.35 and 0.23, respectively. The two exogenous variables together explained 19% of the variance in change fatigue among employees. This outcome supported H1. The more reorganizations, and the greater the impact these reorganizations have on the job, the more employees will suffer from change fatigue.

The second model, with mediating factors included, demonstrated that these factors mediated the direct effect of the intensity of previous reorganizations on change fatigue. However, the mediating factors did not reduce the direct effect of the number of previous reorganizations on change fatigue. That direct effect remained stable. When including the two mediating factors, the total explained variance in change fatigue doubled to 40%, partly supporting H2.

The third model, with mediating and moderating factors included, demonstrated that only one moderating factor—the interaction between number of reorganizations experienced and communication in a prior reorganization—had a significant effect on change fatigue. All the effects of this moderator were negative, suggesting that if communication during a prior reorganization was satisfactory, uncertainty and change fatigue will be reduced. However, including this factor did not reduce the direct effect of the number of previous reorganizations on change fatigue. This effect remained stable and significant (β = 0.36), partly supporting H3. The interaction between satisfaction with communication during prior reorganizations and the number of reorganizations experienced therefore acted as a hybrid moderator. It had the expected effect but added to, rather than replacing, the direct effect of the number of reorganizations on an employee’s level of change fatigue.

H4, in which an additional explained variance was expected by including mediating and moderating factors, was partly supported. The total explained variance in change fatigue increased from 19% to 38% when the mediating factors were included, and it further increased to 43% when the moderating factor was added. This was only the case for one moderator (communication during the prior reorganization) and one mediator (uncertainty). The other factor for which mediating or moderating effects were expected did not show those effects.

Discussion and conclusions

Employees often suffer under repetitive reorganizations. Based on the outcomes of a 2018 survey of employees in a European financial institution, relevant and highly significant relationships were found between the number of reorganizations employees had experienced and their levels of change fatigue (standardized β’s respectively of 0.35 and 0.23, p < 0.01). Employees indicated change fatigue by agreeing with statements about having experienced too many reorganizations, having experienced too many changes to the nature of their work, desiring a period of stability before a new reorganization, and feeling tired of such reorganizations.

The uncertainties related to reorganization, and to a lesser extent the workload, acted as the main generative mechanisms between the number of reorganizations and the level of change fatigue.

Face-to-face interviews and personal observations of employees, with a researcher embedded in the case study organization for over six months, confirmed this outcome. Many employees told the prime researcher that it was about time someone looked into this problem. Some talked about the ‘thousands’ of reorganizations they had been subjected to—an exaggeration, of course, but interesting. In the face of a new reorganization, many employees were less concerned about doing a good job than showing that they were doing an essential job. Many also updated their CVs and frequently discussed the implications of the new reorganization and what it would imply for their department and for them personally.

The structural equation model demonstrated that, to a large extent, our conceptual model was supported. All the effects went in the expected direction, and the model gives a high total explained variance in change fatigue. In the final model, the total percentage of explained variance in change fatigue was a high 43%. The main determinants were the number of reorganizations previously experienced (β = 0.30), the uncertainty involved in reforms (β = .44), the high workload during and after reforms (β = 0.12), and, in a moderating direction, satisfaction about communication during the prior reorganization (β = −0.12) and the interaction between the number of reorganizations experienced and satisfaction about the prior reorganization (β = −0.17).

There was also an unexpected outcome which has theoretical as well as practical implications. The direct effect of the number of reorganizations employees experienced on their level of change fatigue neither disappeared nor weakened when the mediating and moderating factors were included. In fact, the direct effect of the number of reorganizations on change fatigue was neither mediated nor moderated. The factors that we expected to have a mediating impact—workload and uncertainty—did indeed have that effect, but they did not mediate the effect of the actual number of reorganizations the respondents experienced. These factors do, however, mediate the effect of the perceived intensity of prior reorganizations on change fatigue. Including the mediating factors makes the direct effect of the perceived intensity of prior reorganizations on change fatigue disappear. As for most moderating factors, the outcomes are somewhat disappointing. Only one of the expected moderators works in the expected way. Satisfaction with communication during the prior reorganization does help explain change fatigue, but it does not moderate the explanatory power of mediating factors or the explanatory power of the number of past reorganizations on change fatigue.

Our first conclusion is that resistance against organizational reforms is commonly misunderstood. Interpreting reforms in terms of conflicts of interests, conflicts about the goals of the reform, or political, materialistic, or economic issues is incorrect. Experiences with previous reorganizations do impact employees’ state of mind regarding upcoming reorganizations. Whereas the academic literature suggests minimizing resistance by coercive means, eventually crushing it, our research suggests something entirely different. Change fatigue as a form of resistance to upcoming reorganizations emerges not from concerns over the upcoming reform’s intentions, but rather from experiences with earlier reforms. If employees have experienced too many reorganizations with major effects on their day-to-day work, they will be unlikely to welcome a new change. Conceiving change fatigue as simply resistance due to political, materialistic, or economic antagonism therefore misses the point that reorganizations alone have a lasting impact on employees’ mindset. These effects were not at all moderated by successful prior reorganizations or by transformative leadership. Only good communication had some effect in reducing uncertainty and change fatigue.

This conclusion means that each new reorganization has a lower chance of success than the prior reorganization. Regardless of whether the last reorganization worked, the odds of being successful next time decrease. Each time you plan a new major organizational change, change fatigue among personnel increases, resulting in decreasing chances of success.

The urge to constantly reorganize is similar to having a gambling addiction. The advice from statisticians is to do it as little as possible, because gambler’s ruin is lying in wait. The same goes for reorganizations. This research shows that drastic changes should only be a last resort where no alternative solution can be employed. If reorganizations are necessary, the prime concern should be to take measures to avoid and reduce uncertainty among employees, preferably by good communication.

Overall, this research points to a major negative effect of previous reorganizations on the employee mindset. The costs involved in reorganization are often seen as temporary transition costs but would be better seen as structural costs, since reorganizations have a lasting impact on employees.

Disclosure statement

No potential conflict of interest was reported by the author(s).