We would like to thank Detlef Fetchenhauer and Sebastian Lotz for their helpful comments on earlier versions of this manuscript.
This study examines the effect of justice sensitivity on the life satisfaction and job-seeking behavior of unemployed individuals and considers the likelihood of experiencing long-term unemployment. We focus on two facets of dispositional justice sensitivity that reflect individual differences in perception and reactions to perpetrating injustice against others (perpetrator sensitivity) or suffering from the injustice of others as an innocent victim (victim sensitivity). We hypothesised that the negative effect of unemployment on life satisfaction is stronger among individuals with higher levels of victim sensitivity and perpetrator sensitivity. The former are more likely to perceive themselves as victims of an unjust situation, such as fate or the employer's decisions, whereas the latter are more likely to perceive themselves as perpetrators against the rules of social justice. Using survey data from approximately 400 participants, we found that unemployed individuals were less satisfied with life than employed individuals and that this relationship was stronger for perpetrator-sensitive individuals. Unemployed perpetrator-sensitive individuals were more likely to engage in active job-seeking behavior and faced a lower likelihood of long-term unemployment. The results are discussed in terms of the importance of justice-related personality aspects of unemployed individuals for their well-being and labor market outcomes.
Unemployment is a highly stressful life experience (McKee-Ryan, Song, Wanberg, & Kinicki, 2005). Numerous studies have demonstrated that unemployed individuals report lower levels of life satisfaction and happiness (Clark, Georgellis, & Sanfey, 2001; Kassenboehmer & Haisken-DeNew, 2009) and higher levels of depression and other mental health problems (Murphy & Athanasou, 1999; Prause & Dooley, 2001; Winefield & Tiggemann, 1990). Contrary to common belief, the financial difficulties associated with unemployment cannot completely account for the lower satisfaction of unemployed individuals (Blanchflower & Oswald, 2004; Winkelmann & Winkelmann, 1998). Instead, the social costs of unemployment, such as the loss of status and social disapproval, have gained increasing attention from researchers as factors that make the lives of the unemployed so unbearable (Jahoda, 1982; Stavrova, Schlösser, & Fetchenhauer, 2011; Stutzer & Lalive, 2004). Studies that have examined social attitudes toward unemployed individuals and welfare recipients indicate that the myth of the lazy unemployed person is widespread in Europe (Furåker & Blomsterberg, 2003; Golding & Middleton, 1982; Starrin, 2002).
It is argued that justice beliefs are one factor that contributes to the stigmatisation of unemployed individuals. The general public often perceives this group as abusers of the system who violate the rules of social justice by living off others' work instead of working themselves (Furåker & Blomsterberg, 2003; Mansel & Endrikat, 2007; Starrin, 2002; Stutzer & Lalive, 2004). When the unemployed evaluate their situation and react to it, their justice beliefs also play an important role. On the one hand, they often report seeing their “in-group” in the same negative way as the public sees them: as potential cheaters of the welfare state and perpetrators against social justice (Cheek & Piercy, 2003; Kirchler, 1993; Rank, 1994). On the other hand, unemployed individuals have been shown to attribute poverty to structural factors, such as poor economic conditions or society's failure to create enough jobs (Bullock, 1999) and to perceive themselves as innocent victims of unjust fate, especially in cases of structural unemployment (Dalbert, 1997, 2004).
A large body of literature suggests that individual differences in justice beliefs and justice-related personality traits form individuals' judgments, emotional reactions, and behavioral reactions in justice-related situations (Christandl, 2013; Dalbert, 2002; Dalbert & Umlauft, 2009; Lerner & Lerner, 1981; Montada & Schneider, 1989; Schmitt, 1996; Schmitt et al., 2009). Consequently, individual differences in justice perceptions may determine the psychological impact of job loss and behavioral reactions to it. In the present paper, we suggest that individual differences in justice sensitivity—a personality dimension that captures individual differences in the readiness to perceive injustice and the strength of emotional, cognitive, and motivational reactions to injustice (Schmitt, Gollwitzer, Maes, & Arbach, 2005)—shape the reactions of unemployed individuals towards their situation. Importantly, justice sensitivity has been differentiated into two domains: sensitivity to becoming a victim of injustice (victim sensitivity) and sensitivity to the danger of actively committing injustice oneself (perpetrator sensitivity). Individuals scoring high on victim sensitivity react with particularly strong negative emotions to their own undeserved disadvantages or unfair treatment by others, and they tend to perceive a threat of being exploited in ambiguous situations. In addition, heightened sensitivity to being unfairly disadvantaged increases individual rumination after an experience of injustice and inhibits coping (Gollwitzer, Schmitt, Schalke, Maes, & Baer, 2005; Mohiyeddini & Schmitt, 1997; Schmitt & Dörfel, 1999; Schmitt & Mohiyeddini, 1996).
Perpetrator sensitivity captures an individual's sensitivity to injustice for which he or she is personally responsible. Individuals who score high in perpetrator sensitivity are especially sensitive to situations in which they may inflict injustice on others. They react with strong negative self-directed emotions, such as guilt, and they are motivated to restore justice by making amends (Schmitt, Baumert, Gollwitzer, & Maes, 2010).
Based on the assumption that perceptions of injustice play a crucial role in the negative psychological consequences of unemployment, we expect that high (compared to low) scores in victim sensitivity as well as high (compared to low) scores in perpetrator sensitivity lead to particularly low life satisfaction in the unemployed. However, because perceptions of injustice from a victim's perspective and from a perpetrator's perspective should lead to distinct negative emotions and to different motivations, we expect that victim sensitivity and perpetrator sensitivity have different effects on the behavioral consequences of unemployment. As we will explain below, victim sensitivity should decrease motivation to engage in a job search, whereas perpetrator sensitivity should energise the job search and consequently lead to shorter periods of unemployment.
Using the data from a regular pretest of the German Socio-Economic Panel Study (see Haisken-De New & Frick, 1998, for a detailed description), we examined whether individual differences in victim and perpetrator sensitivity determine how much an individual's life satisfaction is affected by job loss. Among unemployed individuals, we studied how victim sensitivity and perpetrator sensitivity affect job-seeking activities and the probability of long-term unemployment.
Unemployment from a Victim's Perspective and a Perpetrator's Perspective
Personality and economic psychology largely address the question of whether different personalities are affected differently by unemployment (for a review, see McKee-Ryan et al., 2005). The literature has suggested a number of personality factors, such as high emotional stability and self-esteem, that buffer the negative effects of unemployment (Creed, Lehmann, & Hood, 2009). Other factors, such as conscientiousness (Boyce, Wood, & Brown, 2010) and neuroticism (Creed, Machin, & Hicks, 1996; Creed, Muller, & Machin, 2001), amplify the negative effects.
Compared to other differences in personality, individual differences in justice perceptions in the context of unemployment have received little research attention. Most of the existing studies assume that the unemployed tend to see themselves as innocent victims and focus on justice-related personality traits that alleviate the negative effects of this view of personal unemployment. For example, in a series of studies, Dalbert and colleagues focused on “belief in a just world” as a personality trait that mitigates the negative impact of unemployment. They showed that individuals who believe in a just world tended to interpret their personal situation as fairly just and did not ruminate about their fate. This strategy, in turn, positively affected their well-being (Dalbert, 1997, 2004; Dzuka & Dalbert, 2002). In another example, Freidl and colleagues have shown that if unemployed individuals appraise their situation in terms of getting what they deserve, the psychological costs of unemployment are reduced (Freidl, Fazekas, Raml, Pretis, & Feistritzer, 2007).
However, studies of public opinion show that the focus on victimisation and undeserved fate represents only one possible view that unemployed individuals can adopt. Studies on public attitudes toward the unemployed indicate that unemployed individuals are more often considered as perpetrators against the norms of social justice than as disadvantaged victims of an unjust fate (Furåker & Blomsterberg, 2003; Montada & Schneider, 1989; Starrin, 2002). Importantly, there is a large body of literature suggesting that the unemployed are aware of the public's negative attitudes towards them and tend to perceive that these attitudes are even more negative than they actually are (Breakwell, Collie, Harrison, & Propper, 1984; Bullock, 1999; Seccombe, James, & Walters, 1998). Most interestingly, unemployed individuals largely share the denigrating attitudes of the general public (Kirchler, 1993). Studies have shown that individuals commonly evaluate their own in-group more positively than other groups, but this positivity bias does not exist among the unemployed (Kirchler, 1993; Sheeran & McCarthy, 1992). Unemployed individuals ascribe negative traits to other unemployed individuals. Rank (1994) reported that approximately 90 per cent of the unemployed individuals he interviewed blamed other unemployed individuals for lacking ambition, for living off of others' work and for cheating the system. Similarly, among welfare-reliant women in the US, attitudes that the unemployed are lazy and abusive of the system have been found (Cheek & Piercy, 2003; Monroe & Tiller, 2001). As these findings suggest, unemployed individuals may be just as likely to take a perpetrator's perspective as a victim's perspective. By determining individual differences in justice sensitivity, it is possible to determine which perspective is more salient for each individual. In other words, we argue that unemployed individuals may potentially perceive that they are undeservedly benefitting from the social system and committing an injustice by not contributing their share to the workforce.
As we will outline in the next section, adopting the perspective of a victim or of a perpetrator is likely to have important consequences for unemployed individuals. Each individual's victim sensitivity and perpetrator sensitivity determine how readily he or she will take on the corresponding perspective and, consequently, are likely to influence the individual's psychological and behavioral reactions to job loss.
Injustice, Well-Being, and Job Search among the Unemployed
Overall, the degree to which unemployed individuals perceive their situation in terms of injustice is important for their well-being. Social justice research provides compelling evidence that subjective injustice fuels negative emotions. Perceiving oneself as an innocent victim of injustice leads to feelings of anger (Mikula, Scherer, & Athenstaedt, 1998; Scherer, Wallbott, & Summerfield, 1986), and cases of prolonged victimisation may lead to helplessness (Kurokawa, 2010; Peterson & Seligman, 1983) and embitterment (e.g. Dalbert, 2011). Perceiving oneself as a perpetrator of injustice triggers self-directed negative emotions, such as guilt and shame (Mikula et al., 1998; Zeelenberg & Breugelmans, 2008), and can lower self-esteem (Tangney, Youman, & Stuewig, 2009). Thus, injustice experienced from both the victim's and perpetrator's perspective can negatively affect well-being and life satisfaction.
However, the motivational tendencies that result from feeling like a victim or a perpetrator of injustice point in quite different directions. Accordingly, unemployed individuals may experience distinct motivations that are crucial for determining whether they actively engage in job search. Angry victims have been found to hold intentions of revenge (e.g. Gollwitzer & Bücklein, 2007) and to focus on the promotion of their own interests (e.g. Batson, Kennedy, Nord, Stocks, Fleming, Marzette, Lishner, Hayes, Kolchinsky, & Zerger, 2007). Similarly, if efforts to actively pursue just rewards are repeatedly frustrated, embitterment not only creates the urge for revenge but also creates despair and passiveness (Linden & Maercker, 2011). These findings suggest that unemployed individuals who feel like innocent victims may not be motivated to invest time and resources in their job search. Rather, it seems plausible that these individuals perceive welfare to be deserved compensation for their disadvantages. On the other hand, unemployed individuals who feel guilty about unjustly profiting from welfare and not contributing to the workforce may be highly motivated to end their unemployment by actively engaging in job search. Feelings of guilt have been found to promote prosocial actions intended to minimise or compensate for the damage inflicted on potential victims (Tangney et al., 2009).
Importantly, the degree to which one perspective, victim or perpetrator, will be adopted by an individual is shaped by dispositional victim sensitivity and dispositional perpetrator sensitivity. Accordingly, if feelings of injustice from a victim's perspective or from a perpetrator's perspective are important psychological processes among unemployed individuals, then victim sensitivity and perpetrator sensitivity will likely determine the emotional and motivational consequences of unemployment.
Victim Sensitivity and Perpetrator Sensitivity
The two justice sensitivity perspectives are not completely independent from each other. A moderate positive correlation has been found between them (r = .33; Schmitt et al., 2010). This correlation means that higher sensitivity from one of the perspectives does not necessarily rule out the possibility of being sensitive to injustice from the other perspective as well. The shared amount of variance for both justice sensitivity perspectives appears to reflect a core concern for justice.
Nevertheless, the justice sensitivity perspectives are correlated differently with external criteria. Perpetrator-sensitive individuals generally score high on other prosocial dimensions of personality, such as tender-mindedness, modesty, straightforwardness (a moral component of agreeableness), and dutifulness, whereas victim-sensitive individuals score high on antisocial personality traits such as Machiavellianism, jealousy, and vengeance (Schmitt et al., 2010; Schmitt et al., 2005). Broadly speaking, victim sensitivity captures quite self-oriented justice concerns, including fear of exploitation, suspicion of others' trustworthiness, reduced willingness to cooperate, and antisocial behavioral tendencies (Faccenda, Pantaléon, & Reynes, 2009; Gollwitzer, Rothmund, Alt, & Jekel, 2012; Gollwitzer, Rothmund, Pfeiffer, & Ensenbach, 2009; Gollwitzer et al., 2005; Rothmund, Gollwitzer, & Klimmt, 2011). In contrast, perpetrator sensitivity represents other-related justice concerns. Individuals who demonstrate high levels of perpetrator sensitivity are characterised by particularly strong emotional reactions to situations in which they potentially inflict injustice on others (Schmitt et al., 2010). They do not exploit situations to promote their own self-interest, and they treat others fairly and altruistically (Lotz, Schlösser, Cain, & Fetchenhauer, 2012; see also Fetchenhauer & Huang, 2004).
Victim sensitivity and perpetrator sensitivity can be expected to play important roles in the psychological consequences of unemployment. Victim-sensitive individuals are more likely to perceive their misfortunes as undeserved and unjust than less victim-sensitive individuals. With regard to reactions to injustice, victim-sensitive individuals have been found to react with strong negative emotions (i.e. anger) towards the perceived perpetrator when they are treated unfairly (Mohiyeddini & Schmitt, 1997). In the organisational context, being treated unfairly has been shown to have particularly negative consequences for highly victim-sensitive individuals in terms of their psychosomatic well-being and the number of sick days taken (Schmitt & Dörfel, 1999). In another study, Schmitt, Rebele, Bennecke, and Förster (2008) reported that among employees who were laid off by their employers, the victim-sensitive employees perceived significantly less procedural justice and placed more responsibility on others compared to the less victim-sensitive employees. Based on these findings, we predict that unemployment may have a particularly negative impact on life satisfaction among victim-sensitive individuals compared to less victim-sensitive individuals (Hypothesis 1).
Furthermore, behavioral reactions to injustice are very likely shaped by victim sensitivity. Victim-sensitive individuals have been found to protest more strongly when they feel unfairly disadvantaged (Mohiyeddini & Schmitt, 1997). In the work environment, victim-sensitive individuals showed more destructive behavior if they perceived unfair treatment by their employers (Schmitt & Dörfel, 1999). Importantly, after being laid off, victim-sensitive individuals showed lower loyalty and a higher tendency towards vengeance against the former employer (Schmitt et al., 2008). Moreover, compared to less victim-sensitive individuals, victim-sensitive individuals were more inclined to behave unethically if they had the opportunity (Gollwitzer et al., 2005) and to make egoistic decisions in social dilemma games (Fetchenhauer & Huang, 2004), presumably behaviors aimed at preventing potential exploitation and receiving compensation for former disadvantages.
These findings suggest that when faced with unemployment, victim-sensitive individuals may feel entitled to live on welfare as compensation for dismissal and may not feel obliged to search for a new job. In addition, victim-sensitive individuals may seek vengeance against the society that condoned the injustice of their unemployment. Consequently, we predict that victim sensitivity is negatively related to active job-seeking behavior (Hypothesis 2). As a natural consequence of our assumption that victim-sensitive individuals are less likely to engage in active job search, we also hypothesise that victim sensitivity is positively related to the duration of unemployment (Hypothesis 3).
Perpetrator sensitivity should provoke quite different psychological processes from victim sensitivity in reaction to unemployment. Perpetrator-sensitive individuals readily perceive that they are inflicting injustice on others (Schmitt et al., 2010). Thus, when unemployed, they may tend to infer that they are committing an injustice against society by living off public funds. For unemployed individuals with low perpetrator sensitivity, this interpretation may be less accessible and less plausible. People with high perpetrator sensitivity are particularly averse to their own unjust behavior. In particular, perpetrator-sensitive individuals react with stronger self-directed negative emotions (i.e. guilt; Schmitt et al., 2010) when anticipating or committing a personal transgression compared to individuals with low perpetrator sensitivity. Because of their heightened readiness to see themselves as perpetrators of injustice, it can be assumed that unemployment triggers negative self-evaluation in perpetrator-sensitive individuals. Therefore, we predict that perpetrator sensitivity amplifies the negative impact of unemployment on life satisfaction (Hypothesis 4).
However, contrary to the impact of victim sensitivity, perpetrator sensitivity may lead to a strong motivation to actively engage in behavior to end the period of unemployment. Perpetrator-sensitive individuals are motivated to prevent personal wrong-doing and to mitigate the potential negative consequences of their behavior. For example, they are more likely than less perpetrator-sensitive individuals to offer an equal split in a dictator game, even when the temptation to behave egoistically is high (Lotz et al., 2012). This evidence indicates that high perpetrator sensitivity increases one's willingness to engage in a particular behavior to prevent injustice, even if this behavior is costly. When they are unemployed, individuals with high perpetrator sensitivity are likely to be highly motivated to stop living on welfare and to re-enter working life to contribute their fair share. In other words, we predict that perpetrator sensitivity is positively related to active job-seeking behavior among unemployed individuals (Hypothesis 5) and, consequently, to lower durations of unemployment (Hypothesis 6).
We utilised data from 455 participants collected in May 2011 during the regular pretest of the German Socio-Economic Panel Study (GSOEP, Innovation Sample). The GSOEP is a panel study that has been conducted since 1984 by the German Institute for Economic Research (DIW Berlin), which conducts annual surveys of approximately 20,000 individuals from approximately 11,000 households. GSOEP questionnaires are pretested annually with a random sample of approximately 1,000 participants who are representative of the German population aged 16 years and older (Siegel, Andreas, & Warnholz, 2009). The SOEP pretest is more comprehensive than the standard protocol for pretesting questionnaires and is often utilised for testing methodological and conceptual questions (Kroh, 2007; Priller & Schupp, 2010; Schonlau, Reuter, Schupp, Montag, Weber, Dohmen, Siegel, Sunde, Wagner, & Falk, 2010). The 2011 pretest that we utilised included interviews with 1,030 individuals (including employed individuals, unemployed individuals, and respondents who are not in the labor force, such as students, housewives, and retired individuals).
Our analyses were restricted to unemployed individuals who were registered at the unemployment office (N = 76) and employed (full- or part-time) individuals, who constituted the reference category (N = 379). The percentage of unemployed individuals in the sample (N = 1,030) was 7.7 per cent, which is similar to the official unemployment rate of 7.1 per cent (OECD, 2011). The mean age was 44.55 (SD = 12.0), and 53.8 per cent of the respondents were female; 29.6 per cent had a university degree, 65.5 per cent had a technical degree, and 14.5 per cent had completed secondary school. Among the unemployed, 71 per cent (n = 54) had been unemployed for 12 months or longer (referred to as the long-term unemployed), and 29 per cent (n = 22) had been unemployed for less than 12 months (the short-term unemployed). A total of 65 per cent (n = 49) were recipients of basic social welfare benefits (Arbeitslosengeld II), 25 per cent (n = 19) received unemployment benefits (Arbeitslosengeld I1 ), and 10 per cent (n = 8) did not receive any financial assistance.
Victim sensitivity and perpetrator sensitivity were measured with brief scales developed specifically for large-scale surveys (Baumert, Beierlein, Schmitt, Kemper, Kovaleva, & Rammstedt, 2012). The response options ranged from 0 (strongly disagree) to 5 (strongly agree). These brief scales included two items to measure each perspective. The items for perpetrator sensitivity were the following: “I feel guilty when I enrich myself at the cost of others” and “It bothers me when I use deceit to achieve something while others have to struggle for it.” The following two statements measured victim sensitivity: “It makes me angry when others are undeservedly better off than me” and “It worries me when I have to work hard for things that come easily to others.” Both scales had good internal consistency (perpetrator sensitivity: α = .83; r(423) = .71; victim sensitivity: α = .79; r(453) = .65). Compared to the population values (taken from Baumert et al., 2012),2 the respondents in our sample had slightly elevated victim sensitivity (M = 3.22 compared to M = 2.95, t(452) = 3.84, p < .001) but did not differ in perpetrator sensitivity (M = 4.18 compared to M = 4.07, t(422) = 1.48, ns).
Life satisfaction was measured using a single item (“How satisfied are you with your life as a whole?”) with a response scale from 0 (not at all satisfied) to 10 (very satisfied). This measure has good external validity (Diener & Suh, 1999; Kahneman & Krueger, 2006), and its convergent validity has been assessed with multiple-item scales of life satisfaction, such as the Satisfaction with Life Scale of Diener and colleagues (Diener, Emmons, Larsen, & Griffin, 1985). This single-item measure of life satisfaction is widely utilised across the social sciences (Boyce & Wood, 2011; Frey & Stutzer, 2002; Lucas, Clark, Georgellis, & Diener, 2004).
To measure job-seeking behavior, the respondents who were registered as unemployed were asked, “Have you actively searched for jobs during the last four weeks?” The answers were coded as 1 if they had and 0 if they had not. The unemployed respondents were also asked to indicate whether they had been unemployed for longer than 12 months (“long-term unemployed”—coded as 1) or for less than 12 months (“short-term unemployed”—coded as 0).
To increase the validity of our results, we controlled for a number of variables in our analyses. We included personality and socio-demographic variables that are known to be related to psychological consequences of unemployment, but that should not be related to perceptions of injustice.
Because prior research has shown that conscientiousness and neuroticism are related to the well-being of the unemployed (Boyce et al., 2010; Creed et al., 1996), we included these as additional control variables related to personality in our analysis. Conscientiousness and neuroticism were measured with brief scales that were designed especially for large-scale surveys; they included three items each (Gerlitz & Schupp, 2005). The scales had acceptable internal consistency despite their brevity (conscientiousness: α = .67; neuroticism: α = .60).
The amount of social support that unemployed individuals receive from their family members or friends is important for their psychological well-being and their behavior in the labor market (Creed & Moore, 2006; Guerrero & Rothstein, 2012; Vinokur & Van Ryn, 1993). Thus, we considered individual differences in social support in our analyses. To measure the degree of social support, we computed an average of three items included in the survey: “How often do you feel that you are lacking the company of other people?”, “How often do you feel as though you are being left out?”, and “How often do you feel as though you are being socially isolated?”(α = .80; response options ranging from 1, very often, to 4, never).
Religiosity has been shown to represent a specific coping technique in situations of uncertainty, threat, and crisis (Hogg, Adelman, & Blagg, 2010; Norenzayan, Dar-Nimrod, Hansen, & Proulx, 2009; Pargament, 1997) and, more specifically, to buffer the negative psychological impact of unemployment (Shams & Jackson, 1993). Consequently, we included individual differences in religious beliefs and practices as a control variable in our analysis. We measured religiosity with an index consisting of the following items: “Independent of any religious denomination, how religious would you say you are?” on an 11-point scale ranging from 0 (not at all religious) to 10 (very religious); “How often do you pray?” on a 5-point scale ranging from 1 (never) to 5 (daily); and “To what degree do you feel connected to your religion?” on a 5-point scale ranging from 1 (not at all) to 5 (very much). The religiosity index had good internal consistency (α = .86).
Because financial assets are an important determinant of life satisfaction in general (Diener, Suh, Lucas, & Smith, 1999) and of unemployed life satisfaction in particular (Feather, 1997), we controlled for whether the respondents possessed financial assets in the form of savings, stocks, or similar resources. The item was “Do you have any assets in the form of savings, stock investments, or similar?”, with yes or no as the response options.
To avoid further confounding variables, we included a number of socio-demographic variables that have been shown to predict life satisfaction in prior research (Diener et al., 1999; Frey & Stutzer, 2002). We controlled for age, gender (1 if male and 0 if female), marital status (dummy categories: married, divorced/separated or widowed, and single as the reference), having children (1 if yes and 0 if no), and education level (dummy categories: a university degree, a technical degree, and no professional degree as the reference). Finally, we controlled for whether the unemployed respondents received any social welfare benefits, including a basic social welfare payment (Arbeitslosengeld II known as HARTZ4), unemployment benefits (Arbeitslosengeld I), or no financial assistance at all.
Table 1 presents the descriptive statistics and zero-order correlations of the variables of interest. The unemployed individuals had lower socioeconomic status than their employed counterparts: they were less likely to possess financial assets, r(435) = −.14, p < .01, and had a lower level of education (only 6% had a university degree compared to 22% among the employed, whereas 30.3% completed secondary school only, compared to 11.3% among the employed).
Table 1. Descriptive Statistics and Zero-Order Correlations
Note: aBased on N = 76; Gender is coded such that 0 = female and 1 = male; Unemployment is coded such that 0 = employed and 1 = unemployed; Financial assets are coded such that 0 = has no assets and 1 = has assets; Children are coded such that 0 = has no children and 1 = has children; Job-seeking behavior is coded such that 0 = no active search and 1 = active search (available only for unemployed individuals).
As expected, unemployed individuals were significantly less satisfied with life than their employed counterparts, r(454) = −.37, p < .001. In line with the results of other studies, life satisfaction was positively and significantly correlated with social support, r(451) = .53, p < .001, conscientiousness, r(453) = .10, p < .05, and material wealth, r(434) = .22, p < .001, whereas it was negatively correlated with neuroticism, r(452) = −.34, p < .001.
There was a positive but relatively weak correlation between perpetrator sensitivity and victim sensitivity, r(423) = .13, p < .01 (e.g. 41.7% of individuals scoring above the median on victim sensitivity had a below-the-median score for perpetrator sensitivity; 47.1% of individuals scoring below the median on victim sensitivity had an above-the-median score for perpetrator sensitivity). Interestingly, the more victim-sensitive individuals were more likely to report lower life satisfaction scores, r(452) = −.16, p < .01, but perpetrator sensitivity was not related to life satisfaction, r(422) = −.02, p = .62.
To test our hypotheses that victim sensitivity and perpetrator sensitivity had different effects on the life satisfaction of employed and unemployed individuals, we conducted a moderated multiple regression analysis with life satisfaction as the dependent variable; perpetrator sensitivity and victim sensitivity, unemployment, and the control variables described in the measurement section; and the products between unemployment and perpetrator sensitivity and unemployment and victim sensitivity.
In Model 1 (Table 2), we entered unemployment, perpetrator sensitivity, and victim sensitivity as predictors. These accounted for 39 per cent of the variance in life satisfaction, F(3, 309) = 18.89, p < .001. As expected, unemployment had a negative effect on life satisfaction, β = −.37, p < .001. Replicating the previous correlation results, the victim-sensitive individuals were less satisfied with life than less victim-sensitive individuals, β = −.14, p < .01, whereas perpetrator sensitivity was unrelated to life satisfaction, β = −.03, ns.
Table 2. The Effect of Perpetrator Sensitivity and Victim Sensitivity on Life Satisfaction of Employed and Unemployed Individuals
Note: Gender is coded such that 0 = female and 1 = male; Employment is coded such that 0 = employed and 1 = unemployed; Financial assets are coded such that 0 = has no assets and 1 = has assets; Children are coded such that 0 = has no children and 1 = has children. List-wise deletion of missing cases (N = 312). We also tested interaction effects of victim sensitivity and perpetrator sensitivity with social support, neuroticism, conscientiousness, and the socio-demographic variables. None of these interaction effects was significant.
To test whether the relationship between perpetrator sensitivity and life satisfaction was different for employed versus unemployed individuals, we introduced an interaction term for unemployment and perpetrator sensitivity in Model 2 (Table 2). The introduction of the product term of unemployment and perpetrator sensitivity led to a significant increase in the explained variance, F(1, 308) = 4.53, p < .05. This result showed that unemployed individuals were indeed more negatively affected if they reported high perpetrator sensitivity (β = −.13, p < .05) compared to the respondents who reported low values.
Figure 1 shows the predicted life satisfaction scores for employed and unemployed individuals with one standard deviation above and below the mean values on perpetrator sensitivity. High perpetrator sensitivity was not related to life satisfaction among the employed, but it substantially reduced the well-being of the unemployed individuals. Using the online tool developed by Preacher, Curran, and Bauer (2006), we computed the confidence bands that identify the range of values of perpetrator sensitivity for which the effect of unemployment on life satisfaction was statistically significant at a 5 per cent level. The estimated bounds of the region of significance were −9.79 (lower bound) and −1.32 (upper bound). Given that the observed values of perpetrator sensitivity in our sample ranged from −3.25 to 1.75, the difference in life satisfaction between employed and unemployed individuals is statistically significant for the participants who scored −1.32 (somewhat less than 1 SD below the mean) and higher on perpetrator sensitivity. Hence, in contrast to perpetrator-sensitive individuals, perpetrator-insensitive individuals do not suffer detrimental psychological effects of unemployment.
To examine whether individual differences in victim sensitivity moderated life satisfaction among unemployed individuals, we added an interaction term between victim sensitivity and unemployment in Model 3. The introduction of the interaction term did not significantly improve the model in comparison to Model 2, F(1, 307) = .02, ns, and the interaction term between victim sensitivity and unemployment did not reach significance, β = .01, ns. Contrary to our expectations, victim sensitivity was not related more strongly to the life satisfaction of the unemployed than to the life satisfaction of the employed.
To examine whether the interaction effect of unemployment and perpetrator sensitivity remained significant when we controlled for the influence of socio-demographics and personality characteristics, in Model 4 we entered the control variables described in the measurement section. The results showed that when the control variables were added to the model, the main effect of unemployment remained significant, β = −.23, p < .001. The main effect of victim sensitivity on life satisfaction was only marginally significant, β = −.10, p < .10. Most importantly, the interaction effect between perpetrator sensitivity and unemployment remained significant, β = −.15, p < .01.
The second goal of our study was to examine whether perpetrator- and victim-sensitive unemployed individuals were more actively engaged in job-seeking behavior than their less justice-sensitive counterparts. A correlation analysis among the unemployed participants (Table 1) showed that victim sensitivity was completely unrelated to the job-seeking behavior of the unemployed, r(73) = .01, ns. By contrast, perpetrator-sensitive unemployed individuals were more likely to conduct an active job search than their less perpetrator-sensitive unemployed counterparts, r(73) = .31, p < .01.
To account for possible confounding variables, we conducted a multiple logistic regression analysis that included job-seeking behavior as the dependent variable, perpetrator sensitivity as the focus predictor, and socio-demographic control variables. In the first step (Table 3), perpetrator sensitivity was entered. It explained 13 per cent of variance (Nagelkerke's R2 = .13), χ2(1) = 7.06, p < .01, and positively and significantly predicted job-seeking behavior, eβ = 1.54, p < .05. That is, unemployed individuals with higher perpetrator sensitivity were more likely to engage in job searches than less perpetrator-sensitive unemployed individuals.
Table 3. The Effect of Perpetrator Sensitivity on Job Search Behavior
Note: Gender is coded such that 0 = female and 1 = male; Financial assets are coded such that 0 = has no assets and 1 = has assets; Children are coded such that 0 = has no children and 1 = has children.
In Step 2, we entered socio-demographic and personality control variables. This led to a significant increase in the percentage of the variance that was explained (Δ Nagelkerke's R2 = .32), χ2(13) = 20.17, p < .10. The results showed that older respondents were less likely to search for new employment than younger people, eβ = .86, p < .01. Divorced or widowed respondents, eβ = 23.04, p < .01, and Arbeitslosengeld II recipients, eβ = 15.32, p < .05, were significantly more engaged in job searches than single individuals and those who did not receive any welfare support from the state, respectively. Above and beyond these effects, the influence of perpetrator sensitivity on job-seeking behavior remained significant, eβ = 1.81, p < .05.
In the following analyses, we examined whether perpetrator-sensitive unemployed individuals comprised a smaller percentage of the long-term unemployed because they were more actively involved in job search. In Table 1, the negative correlation between perpetrator sensitivity and long-term unemployment indicates that perpetrator-sensitive individuals were less likely to be unemployed over the long term than less perpetrator-sensitive individuals, r(73) = −.30, p < .05. Victim sensitivity, however, was not significantly related to long-term unemployment, r(73) = .14, ns.
A logistic regression analysis with long-term unemployment as the dependent variable provided further support for these results (Table 4). The first step included perpetrator sensitivity and explained 12 per cent of the variation in long-term unemployment (Nagelkerke's R2 = .12), χ2(1) = 6.02, p < .05. As expected, perpetrator-sensitive individuals were significantly less likely to remain unemployed for longer than a year compared to their less perpetrator-sensitive counterparts, eβ = .63, p < .05. In Step 2, we entered socio-demographic and personality control variables, which explained an additional 39 per cent of the variance in long-term unemployment (Nagelkerke's R2 = .39), χ2(13) = 24.31, p < .05. The effect of perpetrator sensitivity was robust, eβ = .56, p = .05, indicating that perpetrator-sensitive individuals are less likely to be unemployed over the long term, independent of their socio-demographic or personality characteristics.
Table 4. The Effect of Perpetrator Sensitivity on Probabilities of Long-Term Unemployment
Note: Gender is coded such that 0 = female and 1 = male; Financial assets are coded such that 0 = has no assets and 1 = has assets; Children are coded such that 0 = has no children and 1 = has children.
To examine whether perpetrator-sensitive individuals have a lower probability of long-term unemployment because they are more actively involved in job search, we entered job-seeking behavior in Step 3. This step explained an additional 5 per cent of the variance (Nagelkerke's R2 = .5), χ2(1) = 4.22, p < .05, showing that actively seeking employment is an effective defense against long-term unemployment, eβ = .10, p = .06. The effect of job search substantially reduced the relationship between perpetrator sensitivity and the probability of long-term unemployment, eβ = .68, p = .24, suggesting that job-seeking behavior may mediate the effect of perpetrator sensitivity on unemployment. A subsequent bootstrap test with 5,000 re-samples was used to assess the indirect effect (Preacher & Hayes, 2008). The results showed that the indirect effect of perpetrator sensitivity on long-term unemployment via job-seeking behavior was −.13 (.54). The 95% CI [−2.13; −.01] did not include 0, indicating that differences in job-seeking behavior between perpetrator-sensitive and insensitive individuals transmitted the effect of this personality trait to the probability of long-term unemployment.
In the present paper, we examined the impact of justice-related personality traits, namely justice sensitivity from a perpetrator's perspective and a victim's perspective, on life satisfaction and job-seeking behavior among unemployed individuals.
Our results indicate that victim-sensitive unemployed individuals were not more likely to report lower levels of life satisfaction than their less victim-sensitive unemployed counterparts (Hypothesis 1). We found that victim-sensitive individuals were generally less satisfied with life than less victim-sensitive individuals, regardless of whether they were employed or unemployed. Prior research on victim-sensitive individuals has shown that they tend to have greater perceptions of unfair treatment. For example, victim-sensitive individuals tend to be less satisfied with procedural justice in their organisations than less victim-sensitive individuals (Schmitt & Dörfel, 1999). We can speculate that the perceived degree of unfairness in unemployment for victim-sensitive individuals does not dramatically differ from their perceptions of unfairness and injustices experienced on the job. In other words, victim-sensitive individuals may perceive themselves as undeservedly disadvantaged no matter what their objective life circumstances may be. Engaging predominantly in upward social comparisons and focusing their attention on relative disadvantages, they might neglect the fact that they could be much worse off (and that others are much worse off) than they are. In addition, victim-sensitive individuals might be more likely to react to job loss with agitation, anger, and moral outrage rather than with depressed mood, and these emotions are less closely related to general life satisfaction (Pilcher, 1998).
Although victim sensitivity has been shown to be related to destructive behavior in an organisational context (e.g. Schmitt et al., 2008), our analyses of job-seeking behavior and the duration of unemployment showed no significant differences between victim-sensitive and insensitive individuals in job search activities and the probability of long-term unemployment (Hypotheses 2 and 3). We can speculate that social welfare benefits and re-employment programs may effectively compensate for the perceived unfairness of job loss, prevent victim-sensitive individuals from interpreting their unemployment from a victim's perspective, and thus explain why they did not differ from their less victim-sensitive unemployed counterparts in their job-seeking behavior and duration of unemployment.
Consistent with Hypothesis 4, perpetrator-sensitive unemployed individuals reported lower life satisfaction than their less perpetrator-sensitive unemployed counterparts. This finding is in accordance with a social cognitive approach to personality, which proposes that stable personality traits, such as perpetrator sensitivity, shape an individual's attention to, perception of, and memory of potentially unjust situations (Baumert, Gollwitzer, Staubach, & Schmitt, 2010; Baumert & Schmitt, 2009; Cervone, 2004; Fleeson, 2001). Our results suggest that perpetrator-sensitive individuals were more likely to perceive their joblessness as a personal injustice that they themselves commit. This perception may be because their families have to support them while they are unemployed or because their fellow citizens pay taxes and social insurance that covers the cost of living for the unemployed. This situation can be perceived as an unfair act inflicted on family members or society. This appraisal of one's own unemployment as a violation of the norms of social justice is more common among perpetrator-sensitive individuals than less perpetrator-sensitive individuals and enhances the negative impact of unemployment on life satisfaction.
In addition, as predicted by Hypothesis 5, perpetrator-sensitive individuals were more likely to engage in active job search. Perpetrator-sensitive individuals tend to feel especially badly about their unemployment and should be particularly motivated to end this state of affairs. As a consequence, as predicted in Hypothesis 6, perpetrator-sensitive individuals were less likely to remain unemployed for longer than 12 months compared to less perpetrator-sensitive individuals. In sum, our results suggest that perpetrator-sensitivity is related to positive labor market outcomes.
To our knowledge, the present study is the first to utilise a large and representative dataset to analyze the life satisfaction, job-seeking behavior, and justice sensitivity of the unemployed. The study has methodological advantages and offers a theoretical contribution to the psychological literature on unemployment. First, in comparison to previous studies devoted to justice and the well-being of the unemployed (Dalbert, 1997; Freidl et al., 2007), our analysis of life satisfaction not only includes unemployed individuals but also compares these individuals to employed individuals. This comparison provides an empirical test of our assumption that the influence of justice sensitivity on life satisfaction is specific to unemployed individuals. Second, we showed that perpetrator sensitivity explains the variance in life satisfaction and the job-seeking behavior of the unemployed over and above the impact of various socio-demographic factors and personality traits, such as conscientiousness and neuroticism (Boyce et al., 2010; Creed et al., 1996). Hence, our results cannot be attributed to the confounding effects of other personality traits or socio-demographic variables. In other words, the lower life satisfaction and heightened job-seeking activities of perpetrator-sensitive individuals are driven by their elevated concern for justice rather than other factors. Finally, our results revealed a robust relationship between perpetrator sensitivity and job-seeking activities. The psychological literature suggests that particular personality predispositions determine individual use of coping strategies (Carver, Scheier, & Weintraub, 1989; Terry, 1994). For example, Wanberg (1997) showed that high levels of personal resilience were associated with control-focused coping techniques (such as positive self-assessment and increased control over budgeting). In another research study, Chen and Lim (2012) showed that individual psychological capital (self-efficacy, hope, optimism, and resilience) was positively associated with a specific coping resource—perceived employability—which in turn predicted the probability of problem-focused coping and engaging in job search activities. We speculate that perpetrator sensitivity might also be related to the particular coping strategies that unemployed individuals use. Further research is needed to understand the nature of the relationship between perpetrator sensitivity and coping strategies in the context of unemployment and examine whether this personality trait is also related to success in re-entering the workforce.
Our conclusions are drawn on the basis of data collected using self-report questionnaires. Consequently, there may be a possibility that our estimates are inflated because the covariance between the variables measured with a common method can be partially accounted for by the method itself—a phenomenon referred to as common method variance (Spector & Brannick, 2010). On the other hand, if our findings were biased by common method variance, all self-report variables would show at least a weak correlation. As observed in Table 1, this is far from the case: there are several variables in our study that do not appear to share any common variance with the method used (i.e. they are not correlated—for example, perpetrator sensitivity and life satisfaction overall or social support and religiosity). In addition, recent research has shown that interaction effects in a multiple regression analysis are rather unlikely to be affected by common method bias, and even if they are, common variance deflates the relationships, making them more difficult to detect (Siemsen, Roth, & Oliveira, 2010). Nevertheless, future studies might aim to assess job-seeking behavior using methods other than self-report, such as behavioral observations, to further test the predictive value of justice sensitivity.
Another potential limitation that must be addressed is the relatively small proportion of short-term unemployed individuals compared to long-term unemployed individuals in our sample. This limitation implies that our conclusions are based in large part on the responses of long-term unemployed individuals, which may have slightly biased our findings. Hence, it would be interesting to replicate our findings with samples of predominantly short-term unemployed individuals.
We acknowledge that the correlational nature of our data may weaken our conclusions. However, economic literature on unemployment and happiness based on large-scale panel studies has shown that the main causation effect runs from unemployment to well-being, as assumed in this research, rather than the other way around (Marks & Fleming, 1999; Winkelmann & Winkelmann, 1998). Nevertheless, we are interested in eventually replicating our results with longitudinal data. A longitudinal design will be very helpful in specifying the causal direction in the association between life satisfaction of perpetrator-sensitive unemployed individuals and their job-seeking behavior. Although the analysis of this relationship was beyond the scope of this research, we implicitly assumed that perpetrator-sensitive unemployed individuals suffer from a reduction in life satisfaction and utilise a proactive job search strategy to reduce negative feelings. However, an unsuccessful intensive job search may be even more destructive for well-being than no search at all (Warr, Jackson, & Banks, 1988). This risk could further decrease the life satisfaction of unemployed perpetrator-sensitive individuals. We encourage the utilisation of a longitudinal design in future research, measuring life satisfaction and job-seeking behavior throughout multiple periods of unemployment.
Another interesting research question relates to the cross-cultural generalisability of our findings: does justice sensitivity from a perpetrator's perspective reduce life satisfaction among unemployed individuals globally? We imagine that this effect might be weaker or even non-existent in countries with less extensive systems of social welfare, such as the US. In Scandinavian countries, where social welfare policies are particularly generous, perpetrator-sensitive individuals may be even more likely to interpret their situation as inflicting injustice on the generous state and to have higher motivation to search for new employment. We can also speculate that high tax and social insurance rates, which the unemployed individual paid before his or her job loss, might produce the opposite effect, and the unemployed individual may feel more entitlement to social welfare in these countries.
Even though economic and psychological research has focused on the effect of unemployment on well-being, justice-related personality differences merit more attention. In this paper, we have shown that justice sensitivity from a perpetrator's perspective explains substantial individual differences in life satisfaction and job-seeking behavior among unemployed individuals. Justice sensitivity predicts decisions and behavior in economic games, laboratory studies, and real-life labor market behavior. We hope that our study encourages future consideration of how stable personality differences related to the perceptions of and reactions to injustice explain individual labor market behavior, behavioral outcomes, and choices and attitudes in other domains of life.
Arbeitslosengeld I and Arbeitslosengeld II (as also known as HARTZ IV) are different types of German social welfare benefits. These benefits differ primarily in the source of resource provision (social insurance vs. federal funds) and in the amount paid. Arbeitslosengeld I is paid for 6 to 12 months after job loss and is generally larger than the basic security provided thereafter by Arbeitslosengeld II (HARTZ IV).