The effects of rumors and announcements on employee satisfaction in the context of canceled M&As

(1)

The effects of rumors and announcements on employee satisfaction in

the context of canceled M&As

Nick Assink S3274462

University of Groningen Faculty of Economics and Business

MSc Business Administration – Change Management

(2)

2

Abstract

This research investigated the impact of rumors and announcements of Mergers and Acquisitions (M&As) as antecedents on post-merger employee satisfaction, measured through the ‘turnover of employees’. This was done specifically in the currently under-researched field of canceled mergers and acquisitions, and compared to research focused on completed M&As. Furthermore, it investigated whether being a target instead of an acquiring company has a negative moderating effect on the relationship between a canceled merger or acquisition and post-merger employee satisfaction. In order to investigate these relationships, secondary data from Asset4, Zephyr, and Orbis was collected, and multiple regression analyses were conducted through Stata. The results indicate support for rumors or an announcement of an M&A to have a lasting negative effect on employee satisfaction in case of a canceled merger. It was further found that this negative effect on employee satisfaction was more severe in the case of a withdrawn deal than in the case of a completed merger. Results did not confirm being a target company to have a moderating effect on the relationship between a canceled merger and employee satisfaction. This paper adds to M&A literature by discovering that pre-M&A antecedents negatively and lastingly impact employee satisfaction when an M&A is canceled.

(3)

3

Introduction

Mergers and acquisitions (M&As) are a popular and widespread strategy for companies to grow, diversify and realize growth (Krug & Aguilera, 2005; Uzelac, Bauer, Matzler, & Waschak, 2016). However, it often does not lead to the desired outcomes (Haleblian, Devers, McNamara, Carpenter, & Davison, 2009). Intentions to create synergies do not produce the anticipated benefits (King, Dalton, Daily, & Covin, 2004; Vazirani, 2012), leading to low success- and high failure rates for completed mergers and acquisitions (Appelbaum, Gandell, Yortis, Roper, & Jobin, 2000; Dalton & Dalton, 2008). Synergies are created by combining resources from the involved firms, such as expert knowledge, assets, and economies of scale, in order to gain competitive advantage (Appelbaum et al., 2000a) The high failure rates leave the popularity of M&As unexplained, especially when one considers that the statistics do not account for the many cases of M&A failure before a deal is concluded between companies (Neuhauser, Davidson, & Glascock, 2011).

(6)

6

negatively affects performance, productivity, and resistance (Sanda & Adjei-Benin, 2011). As a result, companies risk the development of employees’ psychological withdrawal and, thus, employee turnover (Fried et al., 1996).

Since a significant proportion of M&A-attempts are eventually canceled, and because this topic is under-researched, this paper will focus on the canceled M&As in order to contribute to current stakeholder theory in an M&A context. It is argued that rumors have a similar effect on employees as announcements, and therefore both will be considered to investigate their influence on employee behavior. Also unknown are the effects of pre-M&A antecedents, such as rumors and announcements, on post-M&A outcomes in the specific cases of withdrawn M&As. The main research question for this study is, therefore, “Does the event of a rumor or announcement of an M&A have a lasting negative effect on employee satisfaction after the withdrawal of the M&A?”.

In order to further investigate the impact of rumors and announcements on employees’ post-M&A emotional states, it is necessary to draw a comparison between withdrawn and completed deals. In this way, this research aims to clarify whether there is a difference in employee satisfaction in cases where an M&A is withdrawn and in cases where the M&A is completed. Therefore, the second research question formulated is: “How does the effect of the M&A announcement or rumor on employee satisfaction after cancellation of an M&A compare to cases where the M&A has been completed?”

Research on the aftermath of completed M&A deals reveals how target companies perceive M&As negatively due to increasing levels of uncertainty, which leads to higher anxiety amongst employees (Ivancevich, Schweiger, & Power, 1987; Appelbaum et al., 2000b). As stated by Reynolds (2015), M&A rumors and announcements contribute to amplification of the effects of uncertainty and anxiety on employees. This leads to the third research question for this study: “Does being a targeted firm have a moderating effect on the relationship between employee satisfaction and a withdrawn merger after an M&A announcement or rumor?”

(7)

7

(8)

8

Literature Review

Mergers and Acquisitions

Mergers and Acquisitions, also referred to as M&As, occur when two organizations willingly agree to join their available assets, liabilities and cultural values on a relatively equal basis across different businesses and industries – called mergers – or when one organization buys and takes over operations from a different organization – called acquisition (Gaughan, 2007; Vazirani, 2012). During this process, the two separate organizations legally cease to exist and are integrated into an entirely new company (Bruner, 2009; Gaughan, 2007). Although M&As are perceived as one of the best ways for firms to realize growth (Uzelac et al., 2016), operational and strategic goals that are set for M&As are rarely met (Haleblian et al., 2009). As described by Appelbaum et al. (2000a): “The primary purpose of merging and acquiring new firms is usually to improve overall performance by achieving synergy” (p. 649). Synergy is defined as the increased economic performance or competitive advantage (Appelbaum et al., 2000a; Appelbaum et al., 2000b; Kinnunen, 2010). M&As are creating such synergy by combining resources of the involved firms. Examples of resources are expert knowledge, assets, or economies of scale or scope (Appelbaum et al., 2000a). However, the potential of synergies is no guarantee that improvements will be realized (Appelbaum et al., 2000a). M&As do not automatically improve the performance of companies involved (King et al., 2004), as M&As maintain to be highly uncertain (Kinnunen, 2010). Moreover, it is stated that a large proportion of M&As end up failing (Dalton & Dalton, 2008), or do not produce the anticipated benefits (King et al., 2004; Vazirani, 2012). Furthermore, a large number of studies show very divergent results in success and failure rates (Appelbaum et al., 2000a; Bruner, 2009; Dalton & Dalton, 2008). Despite these unsatisfactory results in practice, M&As are a frequently used measure to achieve corporate growth and profit objectives (Krug & Aguilera, 2005), leaving the contradiction between their popularity and high failure rates unexplained.

(9)

9

the conclusion of the deal, e.g., the post-merger analyses of Bellinger & Hillman (2000) or Chambers & Honeycutt (2011). The literature remains silent on the repercussions of M&A failure before a deal closure has been reached (Wong & O’Sullivan, 2001). This identifies a research gap in the M&A literature concerning canceled mergers and acquisitions. This becomes especially evident when research on canceled M&As is compared to the amount of research performed on its completed counterparts (Wong & O’Sullivan, 2001).

Canceled or withdrawn M&As (used interchangeably throughout this research) comprise of bids or proposals made to another company, in order to acquire, merge, or be acquired, in both an official or unofficial capacity (Neuhauser et al., 2011). As stated by Neuhauser et al. (2011) , there are many cases of M&A failure before the deal is concluded between companies. This means that proposals concerning intentions to merge, acquiring, or be acquired culminate in a simple failure of this intent. Remarkably, these cases have not received much attention, as a large proportion of takeover bids does not successfully result in an M&A. Authors Holl & Kyriazis (1996) found that approximately 25 percent of their sample on takeover attempts during the 1980s were canceled, while O’Sullivan & Wong (1998) estimate approximately 19 percent of takeover bids in the United Kingdom between 1989 and 1995 were ultimately withdrawn. As such, canceled M&As represent a potential source for a better understanding of the M&A process and its consequences (Wong & O’Sullivan, 2001). The reasons why takeover- and merger attempts are canceled or withdrawn are plentiful. Examples are negative regulatory approval by authorities, defense by target management, rejection by shareholders, or voluntary withdrawal (Wong & O’Sullivan, 2001). Furthermore, pre-merger antecedents and its influence on post-merger behavior and outcomes are also still an under-researched academic avenue (Napier, 1989). This is also argued by King et al. (2004), who state that antecedents of M&A performance have yet to be properly identified. Scholars recognize the importance of the pre-M&A stage while dealing with post-merger outcomes (Quah & Young, 2005), but only a few scholars have been able to use pre-M&A antecedents in order to clarify the controversies regarding the questioned role and impact of M&As (King et al., 2004).

(10)

10

M&As and consequences for Employees

(11)

11

company performance and the value of the organization resources (Chatterjee et al., 1992; Weber et al., 1996; Agrawal & Jaffe, 2000; Meyer, 2001; Very et al., 2005).

As stated, accumulating evidence shows that human resource outcomes are related to business outcomes, and that proper usage of people enhances organizational effectiveness (Koys, 2001). Employee dissatisfaction is also said to influence firm performance and value (Chatterjee et al., 1992; Weber et al., 1996; Agrawal & Jaffe, 2000; Meyer, 2001; Very et al., 2005). This provides a link to stakeholder theory and employee satisfaction. According to Harrison et al. (2015), stakeholder theory promotes “… a practical, effective, and ethical way to manage organizations in a highly complex and turbulent environment” (p. 859). Stakeholders, although subject to differing definitions, are described as individuals, groups and organizations with an interest in the firm, its processes and outcomes, and among which there is interdependence between firm and stakeholder (Harrison et al., 2015). Some of these stakeholders include employees, managers, shareholders, and customers (Harrison et al., 2015). Stakeholder theory, at a minimum, comprises of attending to the interests and well-being of stakeholders (Harrison et al., 2015), in order to create some sort of synergy. This synergy is also known as the concept of generalized exchange (Harrison et al., 2015), which is said to help firms create more value, not only measured in financial terms (Harrison et al., 2015).

Erdogan, Bauer, Truxillo, & Mansfield (2012) formulate life satisfaction as being a key indicator for subjective well-being, relating it to happiness. Satisfaction is defined as the degree of fulfillment provided by life experiences, manifesting itself in all parts of life (Oliver, 1997). Consequently, employee satisfaction is specifically defined as “the feeling that employees have on the jobs; the experience of job in relation to past experience, current expectation, and the alternatives existing in the future” (Hatane, 2015, p. 621). Positive employee satisfaction improves the performance of employees, thereby contributing to the growth of the company (Shaw et al., 2018). Tang & Lee (2014) define employee satisfaction to be based on three main factors: job satisfaction, employee absorption, and employee satisfaction with organizational support. Employee satisfaction is further said to play a prime role in the goal achievements of firms and their performance (Koys, 2001; Chi & Gursoy, 2009). As stated, even the mere announcement or rumor of an M&A already negatively influences employee behavior (Appelbaum et al., 2000a; Appelbaum et al., 2000b; Napier, 1989; Sanda & Adjei-Benin, 2011). The possibility of a future M&A is enough to generate anxiety and initiate survivor-seeking behaviors that negatively affect performance, productivity, and resistance (Sanda & Adjei-Benin, 2011). These survival seeking behaviors, thereby, affect employees’ satisfaction. Survival-seeking behaviors affect employees’ satisfaction, commitment, trust, and intention to stay with the company post-merger, ultimately risking the development of psychological withdrawal and thus, employee turnover (Sanda & Adjei-Benin, 2011).

(12)

12

As discussed above, M&As are still a surprisingly popular measure to realize company growth, despite its high failure rates and consequences on human capital. Oddly enough, current literature does not provide a sufficient explanation for pre-merger antecedents on post-merger outcomes, with canceled mergers still being an under-developed area of research as well. Whereas the strain on human capital is clear for completed cases, little is known about the effects of the pre-merger antecedents (comprising of rumors or announcements) on employees in the cases of canceled M&As. In completed merger cases, this sentiment is described as negative and lasting, which will be tested for the canceled merger scenario as well. Thus, despite still being perceived as growth-inducing for companies, the opposite seems to be true in most cases of M&As. Also, an important consequence of M&As is the negative influence on personnel, which is said to be one of the potential sources of M&A failure. Moreover, within the different stages of an M&A, these consequences have not been studied in a context where the merger or acquisition fails before the actual unification processes have commenced. Furthermore, the M&A in question is said to have negative consequences on employees, leading to the following question: when M&As negatively impact employees, who in turn affect the probability of a success or a failure of an M&A, will the same negative effects remain and endure over time in case of premature termination as well? In other words: will the negative effects of the rumors and announcement prevail or disappear? This leads to the hypothesis:

Hypothesis 1a: The rumors or announcement of an M&A have a lasting negative effect on employee satisfaction in the case of a canceled merger.

As this research seeks to clarify the effects of rumors or announcements on employees in the case of a canceled merger, a comparison to its completed counterparts is necessary in order to test whether these effects are different from one another. M&As’ effects on employees include anxiety and uncertainty. Research does not indicate that cancelation of an M&As terminates these feelings, neither in comparison to its completed counterparts. It can be argued that in completed M&As, less uncertainty about the future exists because of the plans made for the coming merger-process, whereas a canceled merger leaves space for more insecurity about the future. Therefore, the hypothesis:

Hypothesis 1b: The rumors or announcement of an M&A have a more negative effect on employee satisfaction in the case of a canceled merger than in the case of a completed merger.

(13)

13

Appelbaum et al., 2000b). As mentioned before, M&As being an extreme form of organizational change, introduce stress and anxiety to target firms (Appelbaum et al., 2000b). Furthermore, rumors are also said to contribute to target firm employees’ experienced uncertainties (Reynolds, 2015) and increasing anxiety levels (Ivancevich et al., 1987). As a target company confronted with a canceled M&A, this can lead to perceived risk being higher, while predicting the future as a target company could be more difficult than for a acquiring company. Following this logic it can be assumed that this negative impact will furtherly affect the hypothesized negative relationship between pre-merger antecedents and employee satisfaction in the case of a canceled merger as a negative moderator. Therefore, hypothesis 2 is:

Hypothesis 2: Being the target company will negatively moderate the relationship between pre-merger antecedents and employee satisfaction in the case of a canceled M&A.

(14)

14

Methods

Data collection methods

In order to test the hypotheses formulated above, secondary data was derived from different databases, namely, Asset4, Zephyr, and Orbis. Data collection started from the smallest database, Asset4, which is part of DataStream. Here, annual data on companies’ ‘turnover of employees’ (TOE) was downloaded in order to serve as a measure for Employee Satisfaction. This data was retrieved via a static request over the years 1995 till 2017 from the Full Universe List (LAST4ESG). Following, the obtained datasets for the separate years were combined using Stata. After reviewing the combined dataset, the years 1995 till 2001 were dropped from the sample, due to the lack of information on ‘turnover of employees’ for any company listed during these years. Following, the ISIN numbers from this combined dataset were used to search for deals in Zephyr. Here, information on deals concerning the beforementioned ISIN numbers from 01-01-2001 till 31-01-2019 were retrieved, together with information for the control variable ‘industry’. The search included ‘rumored’, ‘announced’, ‘completed-confirmed’ and ‘completed-assumed’ deals in the form of mergers and acquisitions over these years. Following, every deal from Zephyr was split, since every deal consists of an acquirer and target (or multiple targets or acquirers) and listed according to ISIN number. For the research, these were labeled target (t) or acquirer (a), according to their status in the dataset. In the dataset, the rumor-, announcement-, rumor-withdrawn, withdrawn, completed-assumed and completion dates were transformed into years, in order to correspond to the yearly data collected on the turnover of employees. The dates were then divided into two categories: the first category included dates of when rumors started, or when an M&A was announced, depending on which one took place first; the second category included the dates concerning withdrawal or completion of the announcement or rumors.

For the first category, one year was subtracted from each antecedent starting date. This was done to identify the year in which the then coupled data on turnover of employees was still unaffected by the announcement or rumors. These years preceding to the identified dates was computed for all the deals and labeled ‘T0’.

(15)

15

‘T0’ thus accounts for the period before the start of rumors or announcement of a possible deal, and ‘T1’ accounts for the period after a deal has been completed or withdrawn. These years were merged with the corresponding measurements of turnover of employees from Asset4 for both ‘T0’ and ‘T1’. Deals that did not contain measurements on TOE for ‘T0’ and ‘T1’ were eliminated from the sample. This way, a dataset comprising of TOE for the years preceding and following the deals was made. A comparison of this sample was conducted with the initial list of deals downloaded from Zephyr, to account for companies that engaged in multiple deals during the same time period. When a deal from the sample overlapped with the Zephyr deals list for the timeframe between ‘T0’ and ‘T1’, these were excluded to ensure the data on TOE was not contaminated by other deals right before, right after or during the mentioned timeframe. Thereby the possibility of spillover effects was reduced.

Following, each deal was labeled according to their listed status under ‘T1’. For this, ‘completed-assumed’ and ‘completed’ were merged under the category ‘COMPLETED’ and assigned a dummy variable of ‘0’. Then, ‘rumor-withdrawn’ and ‘withdrawn’ were merged under the category ‘WITHDRAWN’ and assigned a dummy variable of ‘1’. This set of data was then unified with the control variables, which are further described below. All observations that did not include information on the control variables were then dropped from the sample.

The industry means and standard deviations were computed as shown in table 1 (Appendix I). Following, the dataset was further adjusted due to the presence of outliers. Therefore, the 10th and 90th percentiles of the data based on the difference in turnover of employees of T0 and T1 were deleted. See table 1 for the values of the percentile-categories before removal. Then, This resulted in a sample of 1740 observations, of which 1630 under the category ‘COMPLETED’, 110 under the category ‘WITHDRAWN’. Under the last category, there were 77 observations that identified as acquirers, and 33 as targets. For ‘T0’ the years range from 2010 till 2017, meaning the rumor or announcement dates range from 2011 to 2018. For ‘T1’ the years ranged from 2012 till 2019, meaning the completed or withdrawn dates range from 2013 to 2018.

__________________________________________________________________________________ Insert table 1 here

(16)

16

Variables

Dependent variable.

The dependent variable for this research is ‘employee satisfaction’. Since employee satisfaction from DataStream was based on company reporting, there is the possibility that these percentages would cover self-reported satisfaction figures, and hence tend to be biased (Berry, Carpenter, & Barratt, 2012). Therefore, another indirect variable that does not contain this bias was chosen. To measure employee satisfaction, datasets from DataStream’s Asset4 Full Universe List (LAST4ESG) on ‘turnover of employees’ (SOEQDP034) were downloaded in the form of a yearly static request. This dataset contains the percentages of employee turnover for the given year of companies. Turnover of employees is said to be an outcome of employee dissatisfaction (Hancock, Allen, Bosco, McDaniel, & Pierce, 2013). This is because M&As bring about dramatic organizational change (Appelbaum et al., 2000b), negatively influencing employee behavior (Appelbaum et al., 2000a; Appelbaum et al., 2000b; Napier, 1989; Sanda & Adjei-Benin, 2011), creating anxiety and uncertainty (Appelbaum et al., 2000b), causing survivor-seeking behavior negatively affecting performance, productivity and resistance and turnover behavior and intentions (Sanda & Adjei-Benin, 2011). A high level of ‘turnover of employees’ corresponds to low levels of employee satisfaction.

For every deal, as described above, both the rumor and announcement dates, and the completion or withdrawn dates, were converted in years and divided into two separate categories, ‘T0’ and ‘T1’. ‘T0’ consists of the year prior to the start of the rumor or the year the announcement was done. ‘T1’ consists of the year after the deal was completed or withdrawn. For these years the corresponding percentages of TOE were listed. This was done to calculate the differences in turnover between the time in which the rumor or announcement did not have an effect on turnover yet (one year before, ‘T0’) and the moment after the merger was withdrawn or completed (one year after, ‘T1’). The difference between these percentages were calculated by subtracting ‘T0’ from ‘T1’ and reported. This thereby measures the effect the rumor or announcement had on the turnover of employees, and thus employee satisfaction.

Independent variable.

(17)

17 Moderator variable.

The moderator variable in this research is whether or not the company in question was a target under the category of the independent variable ‘WITHDRAWN’. This data was collected from Zephyr through the downloading of the information available on deals for the companies from the Asset4 database on turnover of employees. This was done to research the negative moderating effect of being a target company on the relationship of withdrawn M&As and employee satisfaction (measured through TOE). This moderating effect is theorized since research states that employees in target companies experience more uncertainty (Ivancevich, Schweiger, & Power, 1987; Appelbaum et al., 2000b), and that M&A rumors and announcements contribute to amplification of the effects of uncertainty and anxiety on employees (Reynolds 2015). After the download, the deals comprising of both a target and an acquirer were split and separately listed. To this list, the categories of target and acquirer were added in the form of a dummy variable. Targets received the dummy variable ‘1’. Acquirers received the dummy variable ‘0’.

Control variables.

Firm Size. As other researchers who investigated the human side of M&As, firm size is used to

control this research (Cartwright & Cooper, 1993; John, Knyazeva, & Knyazeva, 2015; Luypaert & De Maeseneire, 2015). As Waddock & Graves (1997) this research used ‘Total Assets’, retrieved from the secondary database Orbis, in order to do so. This was done since size of a company might have an influence on turnover on employees, and since firm size has been used as a control variable for strategic change in other research (Zhang & Rajagopalan, 2010)

.

The value of this and all other control variables was taken at time ‘T0’, in order to exclude potential influences on the control variables emerging from completing an M&A or other influences over time.

Industry. In total, the sample has 18 different industries, based on the category ‘primary major

(18)

18

__________________________________________________________________________________

Deal Complexity. Cross-border M&As are said to increase deal-complexity, due to the

combination of companies from different cultures leading to more information asymmetry and increasing costs (Hitt & Pisano, 2004). Therefore, in this study ‘deal complexity’ is controlled for, by determining the differentiation between domestic and cross-border M&As. The sample comprised of ISIN-codes of both the target and acquirer for every deal. These ISIN-codes start with two letters referring to the country where the company in question is based. From the ISIN-codes, the country codes were derived and listed per deal for both the target and acquirer. Then, a comparison was made in order to label them as domestic (same country code) or cross-border (difference in country code). Cross-border deals were assigned the dummy variable ‘0’, and domestic deals were assigned the dummy variable ‘1’.

Firm Performance. The financial performance of a firm can influence turnover of employees,

and is therefore used to control for in this study. For this, two measurements were used.

The first, Current Ratio (CR), also known as working capital (Current Ratio ∑ Current Assets / Current Liabilities) was used to measure how well a company can handle its short-term debt. A ratio with a value below 1 is a warning sign, whereas a ratio equal or higher than 2 is considered to indicate good financial health. This ratio was retrieved from Orbis, and merged with the final sample.

The second measurement used as a control for firm performance is Return on Assets (ROA). When firms do not perform well, turnover of employees is likely to be higher than when a firm does perform well (Hancock et al., 2013). From Orbis, the Return on Asset ratios were downloaded for each observation.

Data Analysis

(19)

19

(20)

20

Results

In this section discusses the results following the multiple linear regression analyses executed to test the relationship between turnover of employees and withdrawn mergers and acquisitions, whether there is a difference in the relationship between withdrawn and completed mergers and acquisitions and turnover of employees, and whether or not being a target has a moderating effect on this relationship.

Descriptive Statistics

(21)

21

is .4962029. The control variable ‘industry’ was determined as described above under ‘control variables’, and consists of 18 dummy variables each representing one of the industry-categories. The minimum for this variable is 1, and the maximum is 18. The mean is 8.151724 and the standard deviation 4.706446.

__________________________________________________________________________________

Correlation Analysis

Table 4 (appendix IV) shows the zero order correlations between the dependent variable ‘turnover of employees’ and the independent and control variables. Zero-order correlations are bivariate correlations that serve as an indicator of relationship strength between two variables, ranging from -1, indicating a perfect negative correlation between two variables, to +1, indicating a perfect positive relationship between two variables. It is important to note that in case of a positive relationship for ‘turnover of employees’, this means that the difference in turnover rate increased between ‘T0’ and ‘T1’, therefore indicating a positive correlation on ‘turnover of employees’, and thus a negative effect on employee satisfaction. The p-values corresponding to these correlations indicate if the correlation between two variables is significant. This research considers a minimum significance level of 0.05. From the table, we can infer that the dependent variable ‘turnover of employees’ is positively correlated with the independent variable ‘M&A completed or withdrawn’ (r= .0499, p ≤ 0.05).

__________________________________________________________________________________

Hypotheses Testing

To empirically test the hypotheses discussed above, the statistical software ‘Stata’ was used. The results can be found in Table 5 (Appendix V). Model 1 describes the regression analysis concerning the dependent variable ‘TOE’ and control variables. Model 2 describes the regression analysis of the dependent variable, control variables and independent variable ‘withdrawn or completed M&A’. Model 3 describes the regression of the independent variables, control variables, independent variables, and the interaction effect ‘target or acquirer’.

(22)

22

Hypothesis 1a predicts a negative relationship between antecedents ‘rumors and announcements of an M&A’ on the post- M&A completion or withdrawal employee satisfaction, measured through ‘turnover of employees’. Therefore, a multiple regression analysis was performed to predict ‘turnover of employees’ from ‘withdrawal or completion of M&A’ and the control variables. The analysis’ outcome for this hypothesis can be found under Model 2 in table 5 (Appendix V). The independent variable and control variables significantly predicted ‘turnover of employees’, F (22, 1717) = 1.85, p <.05, R squared =.0231. The explained variance of this multiple regression model is thus 2.31%. The independent variable ‘withdrawal or completion of M&A’ was significant with β = .3482191 (p < .05). From the zero-order correlations, the relationship between the dependent and independent variable was indicated as positive, meaning an increase in ‘turnover of employees’, indicating a negative relationship in terms of ‘Employee Satisfaction’. Therefore, these results indicate that hypothesis 1a, “The rumors or announcement of an M&A have a lasting negative effect on employee satisfaction in the case of a canceled merger” is supported.

__________________________________________________________________________________

(23)

23

(24)

24

Discussion & Conclusion

Findings

The main aim of this study was to examine whether there were post-merger lasting negative effects on employee satisfaction following rumors or announcements of M&As in the specific case of withdrawal of the M&A. It was further compared to the negative effects known to emerge in the case of completed mergers. It was also researched whether being a target company negatively moderated this relationship. It was found that there is a lasting negative effect on employee satisfaction from rumors or announcements in the case of withdrawn mergers and acquisitions. Furthermore, it was found this effect is more negative in the case of withdrawn M&As than with completed cases. This relationship is not moderated by being a target company instead of the acquirer in the process.

Theoretical Implications

(25)

25

canceled merger, this is a valuable addition. From this research we now understand that canceled mergers have a negative impact on employees and, as consequently can be argued, the company in itself. The psychological impact of M&As is hereby confirmed for canceled M&As, where in this case the impact of rumors and announcements as an antecedent, have a substantial influence on post-merger employee satisfaction. This research furthermore adds to current stakeholder theory in the M&A context. From this study, it can be inferred that attempting an M&A has significant negative effects on the stakeholder category ‘employees’. As stakeholder theory stresses the interconnected relationships between a business and its stakeholders (Harrison et al., 2015), this research shows that the relationship between employees and companies is more negatively influenced in the case of canceled M&As, compared to completed ones. Stakeholder theory also holds the concept of ‘generalized exchange’ (Harrison et al., 2015). Generalized exchange is said to help firms create additional value, also in, but not limited to, financial terms (Harrison et al., 2015). The research shows that this exchange is negatively altered for employees when antecedents such as rumors or an announcement of an M&A arise, thereby influencing amongst others employee loyalty in terms of staying with the firm post-merger. Current research further states increases in shareholder wealth from M&As are results of a re-distribution of wealth from other stakeholder sources (Wong & O’Sullivan, 2001). Proving employee satisfaction is negatively influenced when M&As are withdrawn, this research adds to this perspective in the sense that also in cases of canceled M&As there seems to have been a redistribution of this wealth, causing employee dissatisfaction.

Practical Implications

(26)

26

state, that one of the three main factors of employee satisfaction is ‘satisfaction with organizational support’. Thus, clarifying the rumor or announcement in light of intended organizational objectives, and supporting staff to cope better with the uncertainties of an M&A, can help. Hereby, managers can prevent the discussed survivor seeking behaviors and turnover intentions. By doing so, managers can improve employee retention, decrease consequential losses in human capital, and prevent additional investment in the replacement of employees.

Limitations & Further Research

This research measured employee satisfaction through ‘turnover of employees’. This was done to gauge ‘employee satisfaction’ through an unbiased measure. However, due to reasons related to sample size, it was eventually the only variable measurement considered for this research. However, a problem with the variable ‘turnover of employees’ is that it is not clear whether this turnover has been voluntary or involuntary. Involuntary turnover would not indicate employee dissatisfaction but could indicate, for instance, a strategic effort of a company to lower personnel costs. A better approach would be to combine different variables in order to weigh employee satisfaction, either through the measurement of voluntary turnover of employees or other variables without self-reporting bias.

Another limitation of this research is that ‘type of deal’ was not controlled for. This means that the sample consists of a large variation of M&As, from minority stake transactions to complete mergers or acquisitions. As the intensity of changes within the organization correlates to the perceived uncertainty and anxiety levels with employees, this influences turnover intentions, and consequently turnover rates. Therefore, we can presume that a minority stake transaction correlates less strongly with the ‘turnover of employees’ than a complete M&A does, and that these differences in deals created noise in the sample, making detection of the effects more difficult.

Further limitations of this research are found in the merging of the different categories from Zephyr. Both the categories ‘rumors’ and ‘announcements’ were merged in order to research if there was a pre-merger antecedent that could have influenced post-merger outcomes. Furthermore, the categories ‘withdrawn’ and ‘rumor-withdrawn were merged to ‘withdrawn’, and ‘completed’ and ‘assumed-completed’ were merged to ‘completed’. This leads to the inability to separate the effect of each category on this research, meaning influencing the results described in this paper.

(27)

27

possibly contaminating the data retrieved for those years. Such contamination was accounted for by measuring the difference in ‘turnover of employees’ over the years preceding the rumor or announcement, and the year after the withdrawal or completion. However, it is still arguable that the different dates within a year cause a discrepancy in the data.

Besides the suggestions for future research that can be derived from the described limitations, further suggestions for future research are described below.

As a first suggestion for future research, it could be interesting to investigate how employees perceive the failure of an M&A after this process is terminated, and to compare it to their initial thoughts on the rumors or the announcement of the M&A. An example illustrating why this could be of interest is, that people who in first instance looked negatively at the announcement, can be relieved that the merger was canceled. Another example could be that people who positively appraised an M&A, could perceive its cancellation negatively. Researching these dynamics of human behavior to evaluate whether a certain initial reaction to the merger leads to a specific consequent reaction after M&A cancelation, together with the specific events that influence these reactions, could further contribute to knowledge on this subject. From this, practical information for managers can be derived to guide their employees during the times of uncertainty and anxiety described to be characteristics of M&As.

Another suggestion for future research is to research the dynamics described above in the light of the difference between a target and an acquirer. Even though this research did not show a moderating effect on the specific relationship of hypothesis 1a, it can be supposed that there is still a difference between the target and acquiring firms when it comes to feelings on M&A outcomes.

Furthermore, diving deeper into the subject of pre-M&A antecedents, it is interesting to see what specifically causes employee dissatisfaction with M&As. It could be further researched whether this is based on the antecedents proposed in this research, or for instance, on the lack of communication or other behavior by management.

The next suggestion for future research is to research the different reasons why M&As are canceled, and the influence of such reasons on employee satisfaction or other post-M&A outcomes. Different reasons why mergers are canceled exist: negative regulatory approval; rejection by shareholders; rejection by governments; negative perceptions by the target company; and hostile takeover protection measures (Wong & O’Sullivan, 2001). Investigating these differences could be a positive extension to the research field of canceled M&As.

(28)

28

(29)

29

References

Agrawal, A., & Jaffe, J. F. (2000). The post-merger performance puzzle. In Advances in mergers and acquisitions, 0224(December 1999), 7–41.

Allen, D. (2007). Strategic acquisition amid business uncertainty: Charting course for your company’s M&A. Deloitte Research – Economist Intelligence Unit M&A Survey.

Appelbaum, S. H., Gandell, J., Shapiro, B. T., Belisle, P., & Hoeven, E. (2000). Anatomy of a merger: behavior of organizational factors and processes throughout the pre- during- post- stages (part 2). Management Decision, 38(10), 674–684. https://doi.org/10.1108/00251740010360579

Appelbaum, S. H., Gandell, J., Yortis, H., Roper, S., & Jobin, F. (2000). Anatomy of a merger: behavior of organizational factors and processes throughout the pre- during- post-stages (part 1). Management Decision, 38(9), 649–662. https://doi.org/10.1108/00251740010357267

Belcher, T., & Nail, L. (2000). Integration problems and turnaround strategies in a cross-border merger A clinical examination of the Pharmacia-Upjohn merger. International Review of Financial Analysis, 9(2), 219–234. https://doi.org/10.1016/S1057-5219(00)00034-X

Bellinger, L., & Hillman, A. J. (2000). Does Tolerance Lead to Better Partnering? Business & Society, 39(3), 323–337. https://doi.org/10.1177/000765030003900305

Bellou, V. (2007). Psychological contract assessment after a major organizational change: The case of mergers and acquisitions. Employee Relations, 29(1), 68–88. https://doi.org/10.1108/01425450710714487

Berry, C. M., Carpenter, N. C., & Barratt, C. L. (2012). Do other-reports of counterproductive work behavior provide an incremental contribution over self-reports? A meta-analytic comparison. Journal of Applied Psychology, 97(3), 613–636. https://doi.org/10.1037/a0026739

Bruner, R. F. (2009). Deals from hell.

Buono, A. F., Bowditch, J. L., & Lewis, J.W., I. (1985). When Cultures Collide: The Anatomy of a Merger. Human Relations, 38(5), 477–500.

Capron, L., Dussauge, P., & Mitchell, W. (1998). Resource Redeployment Following Horizontal Acquisitions in Europe and North America ,. Strategic Management Journal, 19(7), 631–661. Retrieved from http://www.jstor.org/stable/3094148

Cartwright, S., & Cooper, C. L. (1993). The Psychological Impact of Merger and Acquisition on the Individual: A Study of Building Society Managers. Human Relations, 46(3), 327–347. https://doi.org/10.1177/001872679304600302

(30)

30

Managing Mergers Acquisitions and Strategic Alliances. https://doi.org/10.4324/9780080510903 Chambers, K., & Honeycutt, A. (2011). Telecommunications Mega-Mergers: Impact On Employee Morale And Turnover Intention. Journal of Business & Economics Research (JBER), 7(2). https://doi.org/10.19030/jber.v7i2.2260

Chatterjee, S., Lubatkin, M. H., Schweiger, D. M., & Weber, Y. (1992). Cultural Differences and Shareholder Value in Related Mergers: Linking Equity and Human Capital. Strategic Management Journal, 13, 319–334.

Chi, C. G., & Gursoy, D. (2009). Employee satisfaction, customer satisfaction, and financial performance: An empirical examination. International Journal of Hospitality Management, 28(2), 245–253. https://doi.org/10.1016/j.ijhm.2008.08.003

Dalton, D. R., & Dalton, C. M. (2008). The New M & A Landscape. Corporate Board, (April), 1–6. Erdogan, B., Bauer, T. N., Truxillo, D. M., & Mansfield, L. R. (2012). Whistle While You Work: A

Review of the Life Satisfaction Literature. Journal of Management, 38(4), 1038–1083. https://doi.org/10.1177/0149206311429379

Fried, Y., Tiegs, R. B., Naughton, T. J., & Ashforth, B. E. (1996). Managers’ reactions to a corporate acquisition: A test of an integrative model. Journal of Organizational Behavior, 17(5), 401–427. https://doi.org/10.1002/(SICI)1099-1379(199609)17:5<401::AID-JOB774>3.0.CO;2-R

Gaughan, P. A. (2007). Mergers, acquisitions, and corporate restructurings (fourth edi). Wiley.

Retrieved from

https://books.google.nl/books?hl=nl&lr=&id=rO230injcpEC&oi=fnd&pg=PR5&ots=-iyGQqyKsg&sig=DTCGRkcW4JKqvoOUL_q1AUSMMHk#v=onepage&q&f=false

Haleblian, J., Devers, C. E., McNamara, G., Carpenter, M. A., & Davison, R. B. (2009). Taking stock of what we know about mergers and acquisitions: A review and research agenda. Journal of Management, 35(3), 469–502. https://doi.org/10.1177/0149206308330554

Hambrick, Donald C, Cannella, A. A. J. (1993). Relative Standing : A Framework for Understanding Departures of Acquired Executives Author ( s ): Donald C . Hambrick , Albert A . Cannella and Jr . Published by : Academy of Management Stable URL : http://www.jstor.org/stable/256757 RELATIVE STANDING : A. Academy of Management Journal, 36(4), 733–762.

Hancock, J. I., Allen, D. G., Bosco, F. A., McDaniel, K. R., & Pierce, C. A. (2013). Meta-Analytic Review of Employee Turnover as a Predictor of Firm Performance. Journal of Management, 39(3), 573–603. https://doi.org/10.1177/0149206311424943

Harrison, J. S., Freeman, R. E., & de Abreu, M. C. S. (2015). Stakeholder theory as an ethical approach to effective management: Applying the theory to multiple contexts. Revista Brasileira de Gestao de Negocios, 17(55), 858–869. https://doi.org/10.7819/rbgn.v17i55.2647

(31)

31

Organization on Financial Performance. Procedia - Social and Behavioral Sciences, 211, 619– 628. https://doi.org/10.1016/j.sbspro.2015.11.081

Hitt, M., & Pisano, V. (2004). Cross-border mergers and acquisitions: Challenges and opportunities. Holl, P., & Kyriazis, D. (1996). The Determinants of Outcome in UK Take-over Bids. International

Journal of the Economics of Business, 3, 165–184.

Huselid, M. A. (1995). THE IMPACT OF HUMAN RESOURCE MANAGEMENT PRACTICES ON TURNOVER, PRODUCTIVITY, AND CORPORATE FINANCIAL. Academy of Management Journal, 38(3), 635–872.

Ivancevich, J. M., Schweiger, D. M., & Power, F. R. (1987). Strategies for Managing Human Resources During Mergers and Acquisitions Human Resource Planning, Volume 10 Number 1 (1987) (despite.pdf. Human Resource Planning, 10(1), 19–35.

John, K., Knyazeva, A., & Knyazeva, D. (2015). Employee rights and acquisitions. Journal of Financial Economics, 118(1), 49–69. https://doi.org/10.1016/j.jfineco.2015.06.001

King, D. R., Dalton, D. R., Daily, C. M., & Covin, J. G. (2004). Meta-analyses of post-acquisition performance: Indications of unidentified moderators. Strategic Management Journal, 25(2), 187– 200. https://doi.org/10.1002/smj.371

Kinnunen, J. (2010). Valuing M & A Synergies as (Fuzzy) Real Options. SSRN Electronic Journal. Retrieved from http://ssrn.com/abstract=1583049

Koys, D. J. (2001). The effects of employee satisfaction, organizational citizenship behavior, and turnover on organizational effectiveness: A unit level, longitudinal study. Personnel Psychology,

54(1), 101–114. Retrieved from

http://onlinelibrary.wiley.com/doi/10.1111/j.1744-6570.2001.tb00087.x/abstract

Krug, J. A., & Aguilera, R. V. (2005). Top Management Team Turnover in Mergers & Acquisitions. Advances in Mergers and Acquisitions, 4(04), 121–149. https://doi.org/10.1016/S1479-361X(04)04005-0

Lipponen, J., Olkkonen, M.-E., & Moilanen, M. (2004). Perceived procedural justice and employee responses to an organizational merger. European Journal of Work and Organizational Psychology, 13(3), 391–413. https://doi.org/10.1080/13594320444000146

Luypaert, M., & De Maeseneire, W. (2015). Antecedents of time to completion in mergers and

acquisitions. Applied Economics Letters, 22(4), 299–304.

https://doi.org/10.1080/13504851.2014.939370

Marmenout, K. (2010). Employee sensemaking in mergers: How deal characteristics shape employee attitudes. Journal of Applied Behavioral Science, 46(3), 329–359. https://doi.org/10.1177/0021886310371373

(32)

32

Perspective. British Journal of Management, 12(1), 47–66. https://doi.org/10.1111/1467-8551.00185

Mylonakis, J. (2006). The perception of banks’ mergers and acquisitions by bank employees. Financial Services Management, 1, 205–214.

Napier, N. K. (1989). M&A’s, human resource issues and outcomes. Journal of Management Studies, 26(3), 271–290.

Napier, N. K., Simmons, G., & Stratton, K. (1992). Communication During a Merger: The Experience of Two Banks. Human Resource Planning, 12(2), 119–136. https://doi.org/10.1007/978-3-322-83820-9_11

Neuhauser, K. L., Davidson, W. N., & Glascock, J. L. (2011). An analysis of failed takeover attempts and merger cancellations. International Journal of Managerial Finance, 7(4), 347–376. https://doi.org/10.1108/17439131111166375

O’Sullivan, N., & Wong, P. (1998). The impact of board composition and ownership on the nature and outcome of UK takeovers. Corporate Governance: An International Review, 6(2), 92–100. https://doi.org/10.1111/1467-8683.00088

Oliver, R. L. (1997). Satisfaction: A behavior perspective on consumer. New York, NY: McGraw-Hill. Quah, P., & Young, S. (2005). Post-acquisition management: A phases approach for cross-border M&As. European Management Journal, 23(1), 65–75. https://doi.org/10.1016/j.emj.2004.12.010 Reynolds, N. S. (2015). Making sense of new technology during organisational change. New

Technology, Work and Employment, 30(2), 145–157.

Sales, A. L., & Mirvis, P. H. (1984). When cultures collide: Issues in acquisition. Managing Organizational Transitions, (January 1984), 107–133.

Sanda, M.-A., & Adjei-Benin, P. (2011). How is the Firm Dealing with the Merger? A Study of Employee Satisfaction with the Change Process. Journal of Management and Strategy, 2(2), 28– 37. https://doi.org/10.5430/jms.v2n2p28

Shaw, J. D., Delery, J. E., Douglas, J. G. J., & Gupta, N. (2018). An Organization-Level Analysis of Voluntary and Involuntary Turnover Author ( s ): Jason D . Shaw , John E . Delery , G . Douglas Jenkins , Jr . and Nina Gupta Published by : Academy of Management Stable URL : http://www.jstor.org/stable/256939 REFERENCES. Academy of Management, 41(5), 511–525. Stahl, G. K., & Voigt, A. (2008). Do Cultural Differences Matter in Mergers and Acquisitions?

Organization Science, 18(1), 160–176. https://doi.org/10.1287/orsc.1070.0270

Stiehl, C., Smith, D., & Omura, A. (2011). After the merger: should executives stay or go? Executive, 4(1), 50–60. https://doi.org/10.5465/ame.1990.4274705

(33)

33

Uzelac, B., Bauer, F., Matzler, K., & Waschak, M. (2016). The moderating effects of decision-making preferences on M&A integration speed and performance. International Journal of Human Resource Management, 27(20), 2436–2460. https://doi.org/10.1080/09585192.2015.1116457 Vazirani, N. (2012). Mergers and Acquisitions Performance Evaluation- A Literature Review. SIES

Journal of Management, 8(2), 37–42.

Very, P., Lubatkin, M., Calori, R., & Veiga, J. (2005). Relative standing and the performance of recently acquired European firms. Strategic Management Journal, 18(8), 593–614. https://doi.org/10.1002/(sici)1097-0266(199709)18:8<593::aid-smj899>3.0.co;2-i

Waddock, S. A., & Graves, S. B. (1997). The Corporate Social Performance-Financial Performance Link Author ( s ): Sandra A . Waddock and Samuel B . Graves Published by : Wiley Stable URL : http://www.jstor.org/stable/3088143 REFERENCES Linked references are available on JSTOR for this article : Strategic Management Journal, 18(4), 303–319.

Weber, Y., Shenkar, O., & Raveh, A. (1996). National and Corporate Cultural Fit in Mergers/Acquisitions: An Exploratory Study. Management Science, 42(8), 1215–1227. https://doi.org/10.1287/mnsc.42.8.1215

Wong, P., & O’Sullivan, N. (2001). The determinants and consequences of abandoned takeovers. Journal of Economic Surveys, 15(2), 145–186. https://doi.org/10.1111/1467-6419.00135

(34)

34

Appendices

Appendix I

Table 1 – 10th_{percentile distribution of ‘turnover of employees’ before adjustment.} Percentile

category (# th₎

N Mean Min Max Sd. Median

(35)

35

Appendix II

Table 2 – Industry list.

Industry N Mean Min Max Sd.

1. Chemicals, rubber, plastics, non-metallic products

292 .1528767 -3.99 4.2 1.634292

2. Construction 76 .2044737 -3.8 4.19 1.96547

3. Education, Health 8 1.035 -1.67 3.5 1.872035

4. Food, beverages, tobacco 53 .485283 -3.71 4.11 2.005649 5. Gas, Water, Electricity 94 -.1589362 -3.9 3.06 1.548647 6. Hotels & restaurants 13 -.1607692 -2.46 3.6 1.628243

7. Insurance companies 2 -1.42 -3.35 .51 2.729434

8. Machinery, equipment, furniture, recycling

347 .0864726 -3.8 4 1.50858

9. Metals & metal products 121 .1317273 -3.48 4.16 1.765465

10. Other services 281 .5013523 -3.51 4.2 1.837014

11. Post and telecommunications 119 .189916 -3.59 4.1 1.577565 12. Primary Sector (agriculture,

mining, etc.)

102 .4954902 -3.7 4 2.064504

13. Public administration and defense 8 .0775 -2.72 2.87 2.0088

14. Publishing, printing 43 .3309302 -3.43 4.14 1.821667

15. Textiles, wearing apparel, leather 4 -.505 -3 2.2 2.133924

16. Transport 59 .3632203 -3.2 4.2 1.714597

17. Wholesale & retail trade 100 .744 -3.6 4 1.643043

18. Wood, cork, paper 18 .1005556 -3.6 3.34 1.5871

(36)

36

Appendix III

Table 3 – Descriptive statistics.

Variables N Minimum Maximum Mean Sd.

(37)

37

Appendix IV

Table 4 – Zero order correlations.

*** correlation is significant at the 0.01 level (2-tailed). **Correlation is significant at the 0.05 level (2-tailed). *Correlation is significant at the 0.1 level (2-tailed)

(38)

38

Appendix V

Table 5 -Regression analyses outcomes.

Variables Model 1 Model 2 Model 3

Deal complexity .099 (.09) .096 (.09) .100 (.09) Total Assets -.000 (.00) -.000 (.00) -.000 (.00) Return on Assets .004 (.01) .005 (.01) .005 (.01) Current Ratio -.063 (.05) -.067 (.05) -.069 (.05) Completed / Withdrawn M&A .348* (.17) Completed M&A .000 (.) Withdrawn M&A . .337 (.20) Acquirer .000 (.) Target -.416 (.30) Completed and Acquirer .000 (.)

Completed and Target .000

(.) Withdrawn and

Acquirer

.000 (.)

Withdrawn and Target .421

(.47) Constant .177 (.15) .165 (.15) .174 (.15)

(39)

39

Appendix VI

Table 6 – Comparison F, R-squared and adjusted R-squared scores.

Column 1 (= completed) Colum 2 (=withdrawn)

F (21, 1608) = 1.64 (15, 94) = 1.63

R squared .0210 .2060

The effects of rumors and announcements on employee satisfaction in the context of canceled M&As