| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||
ORIGINAL ARTICLE |
a Management and Labour Studies Faculty of Business Administration, University of Windsor, Ontario, Canada
Marjorie Armstrong-Stassen, Management and Labour Studies Faculty of Business Administration, University of Windsor, Windsor, Ontario N9B 3P4 Canada E-mail: mas{at}uwindsor.ca.
Decision Editor: Toni C. Antonucci, PhD
Abstract
This panel study examined the reactions of 187 federal government employees aged 45 and older during the initial phase of a large-scale downsizing and 20 months later. There were few significant differences in the reactions of older men and women. Respondents in management positions reported significantly more positive attitudes toward their job and the organization than did respondents in nonmanagement jobs. Compared with the initial phase of the downsizing, respondents reported a significant decrease in commitment to the organization 20 months later. For the two dimensions of job insecurity, perceived threat of job loss decreased, whereas sense of powerlessness over decisions affecting the future of one's job increased. A major area of concern for management is the low level of organizational trust and morale reported by the respondents at both time periods.
IN 1995, the Canadian federal government announced that 45,000 civil service jobs were to be eliminated over a 3-year period. The study reported here is part of a research project designed to assess the long-term impact of the downsizing on those civil servants employed in a department that underwent a large-scale reduction (over 20%) of its workforce during this time. This article focuses specifically on the reactions of remaining employees, that is, the layoff survivors, aged 45 and older to the downsizing. The focus in the research literature has been on the unemployment and reemployment difficulties of older workers who are the victims of downsizing (
Golden 1992
;
Rones 1988;
Warr 1994), whereas the impact of downsizing on those older employees who "survive" organizational downsizing has received little attention.
Why is it important to examine the effect of organizational downsizing on older survivors? There are two major reasons. The first is that older employees may be especially negatively affected by organizational downsizing. Older employees have been more vulnerable to job displacement than younger employees (
Cappelli et al. 1997), suggesting that organizational downsizing may be particularly stressful for this group because of the serious threat of being laid off. Moreover, the practice of offering early retirement incentives as a way of reducing the company's workforce is forcing many older people into involuntarily withdrawing from the labor force (
Useem 1994). People in this age group may also be more reluctant to relocate because of family considerations and a long-term attachment to their community (
Price 1990). Older employees may have a greater vulnerability to being adversely affected by organizational downsizing. In a recent study of the effects of downsizing on the health of remaining Finnish employees,
Vahtera, Kivimaki, and Pentti 1999 found a significant association between downsizing and medically certified sick leave, and this effect was much stronger when the proportion of the workforce aged over 50 years was high, compared with when the proportion aged over 50 was low, suggesting that downsizing may be especially detrimental to older workers. The second major reason for examining the effects of downsizing on older survivors is that organizations have conducted downsizing with little awareness of its age-sensitive effects (
Marshall 1998). It is important that management understands what impact downsizing has on older employees, especially in light of the predictions that older workers will become absolutely critical to businesses in the next few years (
Solomon 1995).
The purpose of this study was to examine the reactions of older layoff survivors to organizational downsizing over a 20-month period and to address the extent to which reactions vary among men and women and among those in management and nonmanagement positions. In their review of the research on organizational downsizing,
Kozlowski, Chao, Smith, and Hedlund 1993 noted that the effects of downsizing may differ on demographic dimensions and, in particular, speculated that survivors' reactions to organizational downsizing may be influenced by their gender and their job level. No research to date has examined the role that individual differences may play in older survivors' reactions.
Hansson, DeKoekkoek, Neece, and Patterson 1997 pointed out the importance of viewing older employees as individuals and the need to address differences among older workers.
The specific reactions I examined included job-related reactions (job satisfaction and job insecurity), organization-related reactions (organizational commitment, organizational trust, organizational morale), and individual well-being (health-related symptoms, burnout). These variables have been identified in the literature as those especially likely to be affected by organizational downsizing. Researchers have found that a workforce reduction may lead to reduced job satisfaction (
Armstrong-Stassen, Cameron, and Horsburgh 1996;
Luthans and Sommer 1999), a dramatic increase in job insecurity (
Allen, Freeman, Reizenstein, and Rentz 1995;
Armstrong-Stassen 1994,
Armstrong-Stassen 1998a;
Cameron, Freeman, and Mishra 1993), reduced organizational commitment (
Allen et al. 1995;
Armstrong-Stassen 1998a;
Jalajas and Bommer 1996;
Luthans and Sommer 1999), reduced trust in the organization (
Cameron 1995;
Kets de Vries and Balazs 1997;
Mishra, Spreitzer, and Mishra 1998), a deterioration in organizational morale (
Cameron et al. 1993;
Jalajas and Bommer 1996;
Jick 1979), an increase in health-related symptoms (
Cappelli et al. 1997;
Fisher 1995;
Vahtera, Kivimaki, and Pentti 1997), and burnout (
Kets de Vries and Balazs 1997). The present study examined whether older survivors would exhibit similar reactions.
This study also investigated whether there are differences in the reactions of older men and women to organizational downsizing. Organizational downsizing may be more stressful for older women, compared with older men, because women may be more vulnerable to being laid off and they experience greater difficulty in becoming reemployed (
Leana and Feldman 1992). For the current generation of older women, most have spent many years either entirely out of the labor force or working only intermittently; thus, they tend to have less seniority than men (
Hale 1990;
Rones 1988;
Shaw 1988). They also may be more restricted than men in relocating to other regions of the country (
Cotton and Majchrzak 1990).
Numerous studies have examined gender differences in job attitudes, although most of these studies have not specifically focused on older workers. The job satisfaction of men and women has been the most frequently studied comparison. The results have been highly inconsistent, with some studies showing that women report significantly higher job satisfaction than men (see, for example,
Clark 1996;
Loscocco 1990;
Tsui and Gutek 1984), some studies indicating that men are more satisfied with their jobs than women (
de Vaus and McAllister 1991;
Miller and Wheeler 1992), and still other studies finding no significant gender differences (
Lambert 1991;
Mottaz 1986;
Neil and Snizek 1988;
Russ and McNeilly 1995).
Abraham and Hansson 1996 compared the job satisfaction of men and women aged 40 to 69 and found that women in this age group reported significantly lower job satisfaction than men. Although women are more vulnerable both to being laid off and to having difficulties becoming reemployed (
Leana and Feldman 1992;
McCall 1997), there is some evidence that men tend to value job security more highly than women (
Rowe and Snizek 1995;
Tolbert and Moen 1998) and are more concerned about, and less satisfied with, their job security than women (
Miller and Wheeler 1992;
Neil and Snizek 1988). However, some studies (
Kuhnert and Palmer 1991;
Roskies and Louis-Guerin 1990) have reported no significant differences between men and women for perceived job insecurity.
The evidence for gender differences for the organization-related reactions included in this study is also inconclusive. For organizational commitment, researchers have found that women expressed significantly higher commitment to the organization than men (
Loscocco 1990), significantly lower commitment than men (
Aven and Parker 1991;
Graddick and Farr 1983;
Marsden, Kalleberg, and Cook 1993), and similar levels of commitment as men (
Aven, Parker, and McEvoy 1993). There has been little research on gender differences in organizational trust. Although there is some indication that, in general, women tend to be more trusting than men (
Johnson-George and Swap 1982), D.
Scott 1983 found no gender differences in trust perceptions of management. No studies were found that have examined gender differences in organizational morale. Because of the inconsistent findings and the scarcity of research, especially in the context of organizational downsizing and restructuring, no predictions were made concerning differences on the job- and organization-related reactions of older men and women.
The evidence indicates that women report experiencing significantly more psychosomatic symptoms and poorer physical health than men (
Davidson and Cooper 1984;
Frankenhaeuser et al. 1989;
Jick and Mitz 1985;
Trocki and Orioli 1994). The research on gender differences in burnout is far less conclusive, and the findings vary depending on the specific burnout component (emotional exhaustion, depersonalization, and personal accomplishment). The present study focused on emotional exhaustion, because this component has been found to be the core component of burnout. Some researchers have found that women report higher levels of emotional exhaustion than men (
Leiter, Clark, and Durup 1994;
Maslach and Jackson 1981), whereas other researchers have found no significant differences between men and women (
Pretty, McCarthy, and Catano 1992;
Williams 1989). Given the inconsistencies in the empirical research on gender differences, it was unclear whether men and women in this older group of survivors would react similarly, or differently, to the downsizing.
Another issue addressed in this article is whether there are differences in the reactions of older employees in management and nonmanagement positions to organizational downsizing. There is evidence that management and nonmanagement employees differ in their attitudes, but again these studies did not specifically focus on older employees. Compared with nonmanagement employees, management-level employees have been found to report significantly higher levels of overall job satisfaction as well as greater satisfaction with various aspects of their jobs (
Carlopio and Gardner 1995;
Mottaz 1986;
Robie, Ryan, Schmieder, Parra, and Smith 1998;
Smith and Hoy 1992), including greater security satisfaction (
Stepina 1985) and higher levels of organizational commitment (
Mathieu and Zajac 1990).
Three studies were found that specifically examined the effect of job level within a downsizing context.
Armstrong-Stassen 1993 found that supervisors were significantly more likely to perceive the company's reason for the cutbacks as fair and were also more likely to perceive greater job security than technicians.
Armstrong-Stassen 1998b examined how layoff survivors perceived and coped with a recent downsizing and found significant differences between female clerical employees and female technicians and between male technicians and male supervisors. In a longitudinal study of the reactions of management and staff employees to a hospital downsizing,
Luthans and Sommer 1999 found that, compared with staff employees, managers reported significantly higher organizational commitment and work-group trust. There were no significant differences between managers and staff employees for job satisfaction. The level of work-group trust remained relatively stable across the three time periods for managers, but trust declined significantly between Time 1 and Time 2 for the staff employees. For both managers and staff employees, there was a decline in organizational commitment and job satisfaction over the 2-year period. These findings indicate that management and nonmanagement survivors respond differently to organizational downsizing. What is not known is whether these differences will continue to be found when older layoff survivors are the focus of investigation.
In summary, this study had three objectives. The first objective was to assess how those layoff survivors aged 45 and older were reacting to the downsizing that was taking place within their department. The second objective was to determine if there are differences in reactions associated with gender and job level within this older survivor group. The third objective was to examine the degree of change in the survivors' attitudes toward their jobs and their organization, as well as in their individual well-being, 20 months later.
Methods
Department Background
The department that participated in this project was targeted for a reduction of over 5,000 employees, representing just over 20% of its workforce. The first step in the downsizing process, which was conducted in late 1995 and early 1996, involved assessing the department's programs and identifying those programs that could be reduced or completely eliminated. The first data collection took place in April 1996, just following the completion of the program review and the identification of the first group of employees to be downsized. This group was offered early departure and early retirement incentive packages and the downsizing that occurred during this period was accomplished through employees voluntarily opting to leave. Employees were aware at this time that subsequent downsizings would be required to meet the targeted reduction, but they did not know which employees would lose their jobs. In the fall of 1996, the department implemented involuntary departures (layoffs) for a second group of employees. The second data collection took place a year later in November 1997. Thus, the first wave of data collection took place during the initial period of voluntary departures, and the second wave of data collection was completed approximately 1 year after the implementation of involuntary departures.
Participants and Procedure
The participants in this study are a subsample of a larger sample randomly selected from the total employee population. I used disproportional stratified random sampling to select the initial sample. All employee names were first stratified by job level (management and nonmanagement) and region (10 regions across Canada and the National Headquarters region in Ottawa). Because the number of employees varied dramatically across the different regions, I used disproportional random sampling to ensure adequate representation for each of the regions.
Each of the selected participants was sent a survey packet in April 1996 (T1) and again in November 1997 (T2). The first survey packet contained a cover letter from the department encouraging participation in the project, a cover letter from the researcher explaining the purpose of the project, and a survey questionnaire. The second survey packet contained a cover letter from the researcher and a questionnaire. All materials were in both official languages (English and French). To match up the T1 and T2 questionnaires, respondents were asked to provide a code number (the first three letters of their mother's first name and the last two digits of their home telephone number). The overall response rate at T1 was 53%. Of those respondents aged 45 or older who completed the questionnaire at T1, 67% completed the T2 questionnaire. The subsample used in this study consists of those respondents who were 45 years of age or older who completed both questionnaires and whose T1 and T2 questionnaire codes could be matched. It was possible to match up the T1 and T2 questionnaires for 187 of the 389 respondents aged 45 and older.
The 187 participants included 98 men and 89 women. Their average age was 48.64 (SD = 3.42, range = 45–67) and they had worked for the department an average of 18.94 years (SD = 6.61). Seventy-eight (43%) were in nonmanagement jobs and 105 (57%) occupied management-level positions (4 missing values). Eighty-two percent were married.
Measures
Unless otherwise indicated, all measures consisted of a 5-point Likert response format.
Job-related reactions
I assessed job satisfaction with eight items, five which were taken from the Job Diagnostic Survey (JDS) developed by
Hackman and Oldham 1974 and three which were developed for this study. Respondents were asked to indicate how satisfied they were with various aspects of their job including the people they work with, the feeling of accomplishment from their job, the amount of challenge in their job, opportunities for advancement, and the amount of work they are expected to do. The Cronbach alpha coefficients at T1 and T2 were .80 and .83, respectively. Job insecurity consists of two dimensions: the perceived threat of job loss and a sense of powerlessness to counteract the threat (
Greenhalgh and Rosenblatt 1984). I measured perceived threat of job loss with three items modified from the Job Insecurity Scale developed by
Jick 1979. These items assessed the degree of worry about one's job, the perceived likelihood of being laid off, and the level of confidence that the department would remain a steady place of employment. A sample item is, "Are you personally worried about your job security?" The alpha coefficients at T1 and T2 were .77 and .81, respectively. Sense of powerlessness was assessed with a single item, "How much influence do you feel you have over decisions concerning the future of your job?" The response categories ranged from 5 (no influence at all) to 1 (a great deal of influence).
Organization-related reactions
I assessed organizational commitment with three items reflecting affective commitment (i.e., emotional attachment to the organization) taken from the Affective Commitment Scale developed by
Meyer, Allen, and Smith 1993. A sample item is, "I feel a strong sense of belonging to [department]." The alpha coefficients at T1 and T2 were .86 and .81, respectively. The three-item measure of organizational trust asked the respondents to indicate how much confidence they had that the organization would look out for their best interests and would always try to treat them fairly. The scale consisted of the two-item trust measure developed by
Ashford, Lee, and Bobko 1989 and an item from the trust scale developed by
Cook and Wall 1980. A sample item is, "I trust this organization to look out for my best interests." The alpha coefficients at T1 and T2 were .85 and .88, respectively. The semantic differential measure of organizational morale consisted of seven pairs of bipolar adjectives designed to reflect enthusiasm for the organization such as enthusiastic–indifferent and encouraged–discouraged. The items were taken from W.E.
Scott 1967. The alpha coefficients at T1 and T2 were .84 and .91, respectively.
Individual well-being
Respondents were asked to indicate how often they had experienced a variety of physical health symptoms (e.g., headaches, pains in the heart or chest, feeling tense or nervous) during the past 6 months. The items were adapted from the Hopkins Symptom Checklist (
Derogatis, Lipman, Rickels, Uhlenhuth, and Covi 1974). The alpha coefficients at T1 and T2 were .90 and .91, respectively. The measure of burnout focused on emotional exhaustion and asked respondents to indicate how often they had experienced a number of job-related feelings (e.g., feeling emotionally drained from their work) during the past 6 months. The six items were from the Emotional Exhaustion subscale of the Maslach Burnout Inventory (
Maslach and Jackson 1981). A sample item is, "I feel burned out from my work." The alpha coefficients at T1 and T2 were .90 and .89, respectively.
Demographic variables
The demographic variables included age, gender, job level, length of time with the department, marital status, and highest level of education completed.
Data Analysis
To determine if there were differences between women and men and between respondents in management positions and those in nonmanagement administrative positions, I conducted general linear model repeated measures analyses of variance (ANOVAs), with gender and job level as the between-subjects factors and time as the within-subjects factor. Because of the number of ANOVAs conducted, I used p < .01 as the significance level instead of the traditional p < .05. To identify significant predictors of T2 reactions, I used hierarchical regression, with the T1 reaction entered in the first step, gender and job level entered in the second step, the T1 job-related reactions entered in the third step, the T1 organization-related reactions entered in the fourth step, and the T1 individual well-being variables entered in the final step.
Results
I first determined if there were any significant differences between those respondents who participated at both T1 and T2 and those respondents who only participated at T1. There were no significant differences for gender, 
2(1, N = 560) = 0.14, p = .70; job level, 2(1, N = 560) = 1.32, p = .25; education, 2(7, N = 559) = 9.11, p = .25; and tenure, t (555) = 0.22, p = .82. I also conducted multivariate and univariate ANOVAs to determine if there were any significant differences between the two groups on the job-related, organization-related, and individual well-being variables. There was no overall significant group effect for the job-related variables, Wilk's lambda F(4,517) = 0.54, p = .71, and the Tukey post hoc comparisons indicated no significant differences on any of these variables. Similarly, there was no overall significant group effect for the organization-related variables, Wilk's lambda, F(3,539) = 0.24, p = .87, and the Tukey post hoc comparisons indicated there were no significant differences between the two groups on any of these variables. There was no significant group effect for health symptoms, F(1,526) = 0.01, p = .94. For burnout, there was a marginally significant difference, F(1,526) = 3.05, p = .08, although as expected the Tukey post hoc comparison indicated no significant group difference. Thus, there is evidence that those who participated at both times did not differ significantly from those who chose not to participate in the T2 study.
The T1 and T2 means and standard deviations by gender and job level are presented in Table 1 and the repeated measures ANOVA F values are shown in Table 2 . In general, the respondents reported relatively low levels of organizational trust, morale, and commitment and a moderately high sense of powerlessness. None of the three-way and two-way interactions was significant. There were only two gender effects significant at the p < .01 level. Women expressed significantly greater satisfaction with their jobs and reported more trust in the organization than did men. There were several significant job-level effects. There were significant differences between management and nonmanagement respondents for job satisfaction, perceived threat of job loss, sense of powerlessness, organizational trust, and organizational morale. Compared with management respondents, nonmanagement respondents reported significantly lower job satisfaction, greater perceived threat of job loss and a greater sense of powerlessness to counteract that threat, and lower levels of organizational trust and morale. There was a significant time effect for perceived threat of job loss, sense of powerlessness, and organizational commitment. Compared with T1, respondents reported a significant decrease in perceived threat of job loss but an increase in sense of powerlessness and a decline in organizational commitment.
|
|
|
For the individual well-being factors, respondents who indicated a greater number of health symptoms at T1 reported significantly less satisfaction with their jobs, lower organizational commitment, trust, and morale, and greater job insecurity and more burnout at T2 than those respondents who reported fewer health symptoms at T1. There were similar findings for T1 burnout, with T1 burnout significantly negatively related to T2 job satisfaction, organizational commitment, trust, and morale, and significantly positively related to T2 perceived job insecurity and health symptoms. In general, higher levels of job satisfaction, organizational trust and morale, and, to a lesser extent, greater organizational commitment at T1 were associated with more positive attitudes toward one's job and the organization and with better individual well-being at T2; whereas higher levels of perceived job insecurity, health symptoms, and burnout at T1 were related to more negative attitudes toward one's job and the organization and to poorer individual well-being at T2.
I ran a series of hierarchical regressions to identify the significant T1 predictors of the T2 reactions. Most of the variance in T2 job satisfaction, organizational trust and morale, health symptoms, and burnout was attributable to the T1 measure of the specific reaction. However, other reactions did explain a significant amount of the variance beyond that of the T1 reaction for the job security and organizational commitment variables, that is, these other variables had a significant effect on the change in job security and organizational commitment over the 20-month period. The results of the hierarchical regression analyses for job security and organizational commitment are shown in Table 4 . T1 health symptoms accounted for a significant portion of the variance in T2 perceived threat of job loss, above that attributed to the T1 measure and job level. Respondents who reported more health-related symptoms at T1 perceived greater threat of job loss at T2. T1 job satisfaction accounted for a significant portion of the variance in sense of powerlessness, beyond that attributable to the T1 measure and job level. Respondents who were less satisfied with their jobs at T1 reported a greater sense of powerlessness about decisions concerning the future of their job at T2. Both T1 job satisfaction and organizational trust accounted for a significant portion of the variance in organizational commitment, beyond that attributable to the T1 measure. Respondents who reported greater job satisfaction and more trust in the organization at T1 reported greater commitment to the organization at T2.
|
This is one of the first studies to examine the reactions of older layoff survivors. There have been studies on older layoff victims—people who have lost their jobs because of downsizing—but older employees who remain in an organization that is undergoing downsizing have received little attention in either the survivor or the gerontological literatures.
Hirshorn 1988 noted that the gerontological research on older workers has ignored the organizational context within which older workers must operate. The survivor literature has emphasized the downsizing context and overlooked the individual characteristics of survivors. The present study is an initial effort to bridge the gaps in the two literatures by focusing on older employees operating within a downsizing organizational context.
Some researchers (e.g.,
Kozlowski et al. 1993;
Leana and Feldman 1992) have suggested that women may be more adversely affected by organizational downsizing because they tend to have less seniority than men and this makes them more vulnerable to being laid off. This should especially be the case for older women who are more likely to have discontinuous work histories. The results of this study show few significant differences in the attitudes of older men and women. Contrary to the speculation of
Kozlowski and colleagues 1993 and
Leana and Feldman 1992, the older women in this study do not appear to have had discontinuous work histories and, in fact, the average length of tenure within the department for men and women was quite similar (19 and 18 years, respectively). Moreover, men were just as likely as women to have been designated for both voluntary and involuntary departure, indicating that for this group of older workers women were not at any greater risk of being laid off than were men. The findings suggest that when older men and women have similar work histories and are equally vulnerable to being laid off, they will react similarly to organizational downsizing.
There were several significant differences between older layoff survivors in management positions compared with those in nonmanagement jobs. Compared with those in management positions, people in nonmanagement jobs reported significantly lower job satisfaction, greater job insecurity, and lower organizational trust and morale. Job level was also a significant predictor of perceived threat of job loss and sense of powerlessness at T2, accounting for a significant portion of the variance in these two variables after controlling for T1 job insecurity. These results are consistent with those of
Armstrong-Stassen 1998b, who found significant differences across job levels in how clerical employees, technicians, and first-level supervisors in a Fortune 100 U.S. telecommunications corporation reacted to organizational downsizing.
The results for the two components of job insecurity are interesting. There was a significant decrease in perceived threat of job loss over the 20-month period but at the same time a significant increase in sense of powerlessness. Compared with the initial phase of the downsizing (T1) which involved voluntary departures, the remaining employees at T2, a year after the implementation of involuntary departures, expressed greater confidence that they would not lose their jobs. On the other hand, their feeling of being able to influence decisions concerning the future of their job declined significantly over the 20-month period. Even when the threat of job loss has diminished, older employees' sense of being able to control their future has been seriously undermined.
Compared with T1, respondents reported a significant decrease in organizational commitment at T2. This is consistent with the findings of other researchers (
Allen et al. 1995;
Armstrong-Stassen 1998a;
Jalajas and Bommer 1996;
Luthans and Sommer 1999) that organizational downsizing results in a deterioration of employees' commitment to the organization. Although other researchers have similarly found a decline in organizational trust and morale following organizational downsizing, in the present study there was no significant change in organizational trust and morale over the 20-month period. However, it is important to note that these older survivors reported a relatively low level of organizational trust and morale during the initial phase of the downsizing and that this persisted across the 20-month period. Even amongst the management-level employees, the average levels of trust and morale were below the midpoint of the scale.
Kanter and Mirvis 1989 found that older workers were more cynical than middle-aged workers, and one of the consequences of this cynicism was less trust in the organization. It may be that a low level of trust and morale existed prior to the downsizing, although it is impossible to determine this given that there was no predownsizing assessment. It is also plausible that the downsizing did adversely affect trust and morale but that this occurred prior to the T1 data collection. There is evidence that organizational downsizing has an immediate and negative impact on trust and morale (
American Management Association 1996), so the damage to trust and morale may have taken place earlier, possibly immediately following the federal government's announcement of the pending downsizing of the civil service in 1995. In either case, from a management perspective, the low level of trust and morale is of concern.
Health symptoms at T1 had a significant effect on the change in perceived threat of job loss over the 20-month period. Health is an issue for older employees, as the evidence shows older people in general report poorer health than younger people (
Rodin and Salovey 1989). In the present study, T1 perceived threat of job loss was significantly positively related to poorer health at T2, and T1 poor health was associated with greater perceived threat of job loss 20 months later. It is possible that this could generate a vicious cycle, with poor health resulting in greater perceived threat of job loss and, in turn, perceived job loss threat leading to subsequent poorer health. This could eventually result in either voluntary or involuntary departure from the organization.
Job satisfaction, which remained relatively stable across the 20-month period, was a significant predictor of the change in sense of powerlessness and organizational commitment. Respondents who reported greater satisfaction with their job at T1 expressed a greater sense of control over decisions affecting the future of their job and a higher level of organizational commitment at T2. The other predictor of T2 organizational commitment that accounted for a significant portion of the variance after controlling for T1 organizational commitment was T1 organizational trust. Respondents who expressed greater trust in the department at T1 reported more commitment to the department 20 months later. These findings suggest that at least some of the negative reactions associated with downsizing will be lessened if attention is paid during the initial period to ensuring satisfaction with one's job and attempting to maintain trust in the organization.
Maintaining trust and morale in an organization that is downsizing is difficult to accomplish, and trying to rebuild shattered trust and morale is a formidable challenge. However, the findings of this study indicate some ways in which this could be accomplished. For example, job satisfaction and sense of powerlessness at T1 were strongly associated with organizational morale and trust at T2. Employees who were more satisfied with their jobs and who felt that they had greater input into decisions concerning the future of their jobs during the initial phase of the downsizing reported higher organizational trust and morale at T2. This suggests that the organization may be able to improve trust and morale by taking measures to improve employees' job satisfaction and by providing employees greater input into decisions concerning the future of their jobs.
The strengths of this study are the focus on older workers and the longitudinal research design. Although the findings are based on self-reported data, the fact that the data were gathered 20 months apart makes the issue of common method variance less problematic. There are, however, limitations to the present study. The research project was designed to examine the reactions of survivors in general, not just older survivors. Therefore, variables that may be especially relevant for older individuals may have been omitted. There is also a question of generalizability of the findings. All of the respondents were federal government employees and the findings may only generalize to that population. Moreover, the work history of the older women in this study may not be representative of the work histories of older women in general. The evidence does suggest that many women in this age group tend to have discontinuous work histories and are therefore more vulnerable to layoff as
Kozlowski and colleagues 1993 suggested. The small number of people in some of the subgroups (e.g., there were only 29 men in the nonmanagement group) may have contributed to the nonsignificant results for the Gender x Job Level x Time interactions.
The findings of this study make an initial, yet important, contribution to our knowledge of the impact of organizational downsizing on older employees—a group that to date has been ignored in the survivor literature. Clearly, this study needs to be replicated with a much larger, and more representative, sample. It is important that researchers turn their attention not only to investigating the impact of downsizing on older employees but also to identifying those older individuals who are most vulnerable to being adversely affected by organizational downsizing.
Received for publication December 2, 1999. Accepted for publication July 27, 2000.
References
| ||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
