The Global Financial Crisis and Individual Distress: The Role of Subjective Comparisons after the Collapse of the Icelandic Economy

Abstract

Sociologists have emphasized that abrupt social changes can evoke subjective deprivation that can create subjective injustice and emotional distress. The global economic crisis offers an opportunity to examine this issue. This article builds on relative deprivation theory and proposes that economic crises evoke three subjective comparisons that influence distress: comparisons to 1) past outcomes; 2) the situation of others; and 3) expected future outcomes. Using a national survey obtained during the economic crisis in Iceland, we examine how these comparisons influence subjective injustice and emotional distress (anger and depression). Results indicate that perceived reduction in the standard of living has a more pronounced effect on subjective injustice and anger, 1) when individuals think that the crisis has harmed them more than others; and 2) when they have negative expectations about their future. The study implies that subjective comparisons can moderate the effect of sudden social change on distress.

Keywords

economic crisis emotional distress Iceland relative deprivation subjective comparisons subjective injustice

Introduction

Sociologists have emphasized that abrupt social changes can trigger subjective deprivation that can create subjective injustice and emotional distress. Economic fluctuations provide a case in point. As Durkheim (1951[1897]) argues in his seminal work on suicide, sudden economic fluctuations may change social conditions so rapidly that they become temporarily inconsistent with common goals that people have internalized. At such historical moments, many feel unable to achieve their goals and may experience injustice and frustration. In a similar vein, Davies (1962) argues that in an economic downturn many people experience injustice and frustration because they no longer have things that they value and had in the past. For Davies, a crisis is particularly conducive to injustice and frustration if it occurs suddenly after a prolonged period of increasing prosperity, because during such periods people internalize high expectations about their economic future that are then shattered in the crisis.

In short, theorists have emphasized that sudden social changes, including economic crises, can create distress, because not only can they create economic problems, but they may also trigger subjective deprivation when people suddenly face a changed world. However, limited direct research exists on how subjective deprivation may link crises and individual distress. Studies have examined aggregate-level effects of economic downturns on rates of physical and mental health problems, substance use, suicide, and crime (Brenner, 1987; Durkheim, 1951[1897]), and individual-level research has examined the effect of objective hardship on emotional distress during recessions, including unemployment and economic hardship (Dooley et al., 1994; Tausig and Fenwick, 1999). Research generally finds harmful effects of recessions on mental health, although there is heterogeneity in the findings (Zivin et al., 2011). Thus, research indicates that crises generally increase distress, but since it has mostly focused on the (aggregate- or individual-level) effects of economic factors (see Krahn and Harrison, 1992), we know little about the mechanisms through which crises influence distress (see Hedström and Swedberg, 1998).

Relative deprivation theory provides a starting point to identify some of these mechanisms. Relative deprivation refers to the perception of unfair disadvantage compared to a reference point that is salient to the individual; for example, compared to others in society, or to one’s own situation at another point in time (Crosby, 1976; Merton and Rossi, 1968; Runciman, 1966; Smith et al., 2011; Stouffer et al., 1949). In the current article, we build on relative deprivation theory and argue that economic crises evoke and/or render three types of subjective comparisons more salient. First, people compare their current economic situation to what it was before the crisis (comparison to past outcomes). Second, they evaluate whether the crisis has harmed their economic situation more than it has harmed the situation of others in society (comparison to the situation of others). Third, crises encourage people to anticipate their future situation (comparisons based on expected future outcomes). We argue that these three comparison processes interact in influencing subjective injustice and emotional distress during economic crisis.¹

Large-scale crises, such as the recent global economic crisis, provide opportunities to study processes linking abrupt social change and individual experience. Hit particularly hard and abruptly by the economic crisis, Iceland provides an ideal social setting for such research (Benediktsdóttir et al., 2011). As in other affluent countries, a prolonged period of economic growth preceded the economic crisis in Iceland. This period of rising expectations ended abruptly in a national drama in a single week in October 2008, in a sudden bankruptcy of almost the entire banking system in Iceland. The subsequent crisis entailed a large drop in living standards and widespread economic difficulties and rising unemployment. Moreover, the crisis triggered widespread discontent among the Icelandic public that entailed the perception that the collapse had resulted from corruption in the financial and political sectors.

The current article uses Iceland as a test case for examining the role of subjective comparisons in distress during an economic collapse. In what follows, after describing the social context of the Icelandic crisis, we build on relative deprivation theory and discuss the three comparison processes presumably evoked by major economic crises. We then use a national survey conducted during the midst of the economic crisis in Iceland, in 2009 and 2010, and examine how the comparison processes influence subjective injustice and emotional distress (anger and depression).

The Economic Crisis in Iceland

In the years leading to the financial collapse in 2008, between 1993 and 2007, growth characterized the Icelandic economy and the public experienced large increases in the standard of living, triggering rising goals and expectations. Many families bought new and/or bigger houses, renovated their houses with expensive materials and equipment, bought multiple new expensive cars, and travelled abroad several times a year. Many of those who did not immediately achieve such goals developed expectations that they would attain them at some point. This development intensified in the years leading up to the crisis, following the privatization of two large state banks in 2002. Between 2004 and 2007, annual economic growth averaged around 6.5 percent. The banks enhanced the public’s goals and expectations by offering easy loans to buy cars and houses, and the press dutifully reported on the lifestyle of the rich and the consumerism characterizing their way of life. In 2007 Iceland reportedly had the highest standard of living in the world (United Nations Development Programme, 2007), and opinion polls showed a historical peak in the public’s economic expectations.² Icelanders had become used to thinking that they would continue to enjoy rising economic affluence.

The financial collapse thus created an abrupt change in Icelandic society. The three largest banks in Iceland, all former state banks, went bankrupt in a period of nine days in October 2008. As these banks jointly comprised about 85 percent of the banking system, the economic shock was severe (Matthíasson, 2008). The collapse almost wiped out the Icelandic stock exchange completely, with huge losses for investors as well owners of capital and real estate (including many homeowners). The Icelandic currency fell more than 95 percent against the dollar, and Iceland became the first developed country to require assistance from the International Monetary Fund in 30 years. National economic production decreased by 6.7 percent and 4 percent in 2009 and 2010, respectively (Statistics Iceland, nd.a), and private consumption may have fallen by more than 20 percent during autumn 2008 and spring 2009 (Jónsson, 2011). Unemployment increased from 1 percent of the labour force in 2007 to 8 percent in 2009 (Statistics Iceland, nd.b).The Icelandic public experienced media coverage of a bleak economic future for the country, including the threat of the state not being able to pay its foreign debt. During the recession in 2009 and 2010, expectations about the country’s economic affluence reached a historical low point in opinion polls.³

In the minds of many Icelanders, corruption caused the economic crisis. Prominent public voices stated that the bank crash resulted from corruption in the financial sector, stemming in part from a failure of the government to regulate it (Report of the Special Investigation Commission, 2010). The widespread public discontent created a legitimacy crisis for Icelandic politics, which intensified in a series of public meetings held in downtown Reykjavík during fall 2008 and ultimately resulted in mass protest and riots that brought down the ruling government in January 2009.

Subjective Comparisons and Relative Deprivation

Relative deprivation theory emphasizes that subjective comparisons influence how people experience their situation, and that they do so independently of the objective attributes of the situation (Crosby, 1976; Merton and Rossi, 1968; Runciman, 1966; Stouffer et al., 1949). How individuals perceive their situations in comparison to salient reference points shapes their level of satisfaction, and thus they do not need to experience objective economic hardship to be dissatisfied. Perceiving an outcome unfavourably in comparison to a salient reference point, that is, having negative ‘referent outcomes’ (Folger, 1984), may create feelings of relative deprivation (Crosby, 1976). Such feelings of unfair disadvantage are presumably distressing and harm well-being. These can be feelings of having an unjust status or they can appear as emotional distress manifested through anger and depression (Smith et al., 2011).

Various reference points are potential sources for subjective comparisons, including 1) one’s past outcomes; 2) perception of the situation of others; and 3) expected future outcomes (Levine and Moreland, 1987). First, past outcomes provide a reference point in that they have shaped current wants and expectations. Second, the perception of other individuals or groups with whom individuals identify with influences how they experience their situation, because such comparisons shape their goals, expectations, and ideas about deservingness (Runciman, 1966). Thus, individuals can experience relative deprivation when thinking that their situation is worse than that of others with whom they identify and compare themselves (Bernburg et al., 2009; Gartrell, 2002; Merton and Rossi, 1968). Third, expected future outcomes constitute a reference point. Individuals are less inclined to accept a perceived disadvantage if they think that this situation will not improve or will become worse, while they are more likely to accept a disadvantage if they anticipate an improvement (Levine and Moreland, 1987).

Comparisons in the Social Context of an Economic Crisis

Building on the general idea that social context shapes subjective comparisons (Runciman, 1966), we propose that major economic crises can evoke and/or render salient three types of subjective comparison: comparison to past outcomes, to the situation of others, and to expected future outcomes.

Comparison Based on Past Outcomes

The historical-social context influences how individuals select their reference points (Graziano, 1987; Mirowsky, 1987). Thus, a sudden economic crisis may direct the individual’s reference point to a better past. Many individuals experience a drop in their standard of living, creating a negative comparison to their past situations. Such comparisons may trigger relative deprivation, especially if the crisis occurs suddenly after a prolonged period of prosperity during which individuals have internalized high expectations (Davies, 1962). Providing an ideal social setting for examining this idea, Iceland went through a prolonged period of economic prosperity and rising expectations that ended abruptly in the financial collapse.

Moreover, social context shapes the tendency for perceived disadvantage to result in feelings of unfairness (Crosby, 1976; Markovsky, 1985). Individuals tend to evaluate perceived disadvantage as unfair if the social context defines it as the result of illegitimate processes (Lind and Tyler, 1988; Skitka, 2009). Also, people are more likely to evaluate a perceived disadvantage as unfair if social context encourages them to blame it not on themselves but on forces beyond their control (Crosby, 1976; Reis, 1987; Testa and Major, 1990). Accordingly, perceived loss blamed on an economic crisis may be particularly likely to trigger relative deprivation, as individuals may tend to blame their perceived loss on forces beyond their control (e.g. inflation or cutback in public welfare). Moreover, as in the Icelandic crisis, people may see their reduction in the standard of living as caused by illegitimate processes, such as corruption, further triggering relative deprivation. Again, it is a widespread view among the Icelandic public that corruption caused the crisis.

Few studies have examined how perceived reduction in standard of living during an economic recession influences distress. A survey conducted by Krahn and Harrison (1992) during a 1980s recession in Alberta, Canada, is an exception. Building on relative deprivation theory, this study found that working-class people who perceived a reduction in their standard of living in the recession indicated more support for government redistribution, suggesting that experiencing a reduction in living standard in a crisis can trigger grievance toward the social order.

Accordingly, we hypothesize that individuals who perceive a negative change in their standard of living during the crisis will experience more subjective injustice, anger, and depression, net of their current economic situation (Figure 1, path a).

Figure 1.

Hypothesized effects of subjective comparisons on distress (all paths have a predicted positive sign).

Levine and Moreland (1987) have argued that individuals can have several salient reference points simultaneously, suggesting an interplay (interaction effects) among different comparisons. We propose that two other comparison processes may moderate the effect of perceived reduction in standard of living on distress: 1) the perception of how the crisis has harmed others; and 2) expectation of future outcomes.

Comparison to the Situation of Others

Social theorists (Merton, 1968; Runciman, 1966) have emphasized that democratic features, such as equal opportunity, egalitarian values, and high social mobility, encourage people to compare themselves to common standards, as opposed to restrict their comparison to individuals in their personal networks or social class. Democracies promote unrestricted comparisons, whereby people tend to evaluate the justice of their situations in reference to generalized ideas about the situations of others in the society.

A sudden, large-scale crisis may lead individuals in democratic societies to evaluate whether the crisis has harmed them more than other members of the society. The Icelandic context in particular is likely to encourage individuals to compare whether the crisis has harmed them more than other Icelanders. Bernburg et al. (2009) have argued that, due to the relative ubiquity of social mobility, shared belief in equal opportunities, and cultural homogeneity, Icelandic society promotes unrestricted social comparisons based on national identity. Moreover, as Tajfel (1978) has argued, when a group collectively experiences an adversity this same group can become a stronger reference group for its members. The economic collapse was particularly dramatic, a collectively-felt event that may have strengthened Icelanders as a reference group.

Given the pivotal role of social comparisons in the evaluation of disadvantage, we suspect that people may evaluate the justice of change in their living standard due to crisis based on such comparison. That is, individuals may evaluate the justice of their perceived economic loss due to crisis based on how, in their minds, the crisis has harmed their situation relative to how it has harmed the situation of others in the society. In what follows, we refer to this experience as the perceived relative impact of the crisis.

This argument implies an interaction effect, whereby one comparison process (comparison with the situation of others) conditions the effect of another (comparison to past outcomes) on distress. Thus, the perceived relative impact of the Icelandic crisis may modify the association between reduced living standard and distress (Figure 1, path b). Even a large drop in the standard of living during crisis may not trigger much distress if individuals think that the crisis has harmed others more than themselves. Conversely, even a small drop in the standard of living may create distress if individuals believe that the crisis has harmed them more than most others. The perceived relative effect of the crisis may also influence distress directly (Figure 1, path d).

Comparison Based on Expected Future Outcomes

Economic crises can create widespread anxiety about the future (Brenner, 1990). In a recession, fearing job loss, inflation, cuts in public welfare and so on, expectations about one’s future may become a particularly salient reference point. Research has shown that expectations about a worsening future status, such as job loss, can harm health and mental well-being (Burgard et al., 2009). Crises provide an opportunity to examine how expectations influence distress. We test whether negative expectation of the future influences subjective injustice and emotional distress (Figure 1, path e). Moreover, expectations may condition the individual’s experience of economic loss, implying an interaction effect. Specifically, perceived reduction in living standard may have a more pronounced effect on distress when individuals have negative expectations about future outcomes (Figure 1, path c).

Relative Deprivation and Individual Outcomes

Research has underscored that relative deprivation – a feeling of unfair disadvantage created by negative comparison – can create distress (Smith et al., 2011). Thus, unfavorable comparisons can trigger perceived injustice and anger (Bernburg et al., 2009; Testa and Major, 1990), depression and anxiety (Smith and Ortiz, 2002), and worse physical health (Yngwe et al., 2005). We focus on three outcomes that should stem from negative subjective comparisons: perceived injustice, anger, and depression. Given well-known gender differences in emotional distress (Rosenfield and Smith, 2010), it is important to capture both internal (depression) and external (anger) expressions of distress. This strategy allows us to examine the robustness of the results across different outcome variables.

Method

Face-to-face interviews were conducted using a national sample of Icelandic adults (18 years and older). Trained interviewers visited all participants in their own homes. For probability sampling, we used a complete list of all individuals who have an Icelandic social security number (Iceland’s National Registry). We used cluster sampling, to save cost in obtaining data outside of the metropolitan area. We created 17 clusters representing differently populated areas nationwide, and then randomly selected individuals from each cluster. The number of individuals selected randomly from each cluster was directly proportional to the population living in areas represented by the cluster, so weighting is not necessary.⁴

The final response rate is 63 percent (948 interviews completed). Deletion of cases due to missing values resulted in 812 valid cases for the current analysis (86 percent of completed interviews). We have examined how the sample represents selected population parameters: gender, age, and employment. The sample reflects these parameters quite well, although males are slightly under-represented (constituting 49 percent of the sample but 50 percent of the population), as are young people (21 percent of the sample are individuals between 18 and 29 years of age, a group constituting 24 percent of the population).

Dependent Variables

We use an index for subjective status injustice comprising the mean score on two, five point questions: ‘Do you ever get angry or frustrated due to your status in the society?’, and ‘Overall, do you feel that your current status in the society is just or unjust?’ The index ranges from 1 (minimum subjective injustice) to 5 (maximum subjective injustice). Cronbach’s alpha is 0.63. Anger is measured with the mean answer for five questions about how often during the week prior to the survey respondents experienced irritability, fits of temper, arguments, a want to break or smash things, and whether they had screamed or thrown things (see Bernburg et al., 2009). The answers ranged from 1 (never) to 4 (very often). Cronbach’s alpha is 0.70. Finally, we measure depression using the mean score on three questions about how often during the week prior to the survey the respondent experienced ‘sadness or having little interest in doing things’, ‘feeling sad or blue’, and ‘not being excited about anything’. Answers ranged from 1 (never) to 4 (very often). Cronbach’s alpha is 0.81.

Subjective Comparisons

We created three novel measures for the subjective comparison processes: perceived change in standard of living during the crisis, perceived relative financial effect of the crisis, and expectations. Perceived change in standard of living is a scale: ‘Compared to the way your standard of living was before the crisis hit in fall 2008, how is your standard of living today?’ The point scale ranges from 1 (much better than before the crisis) through 4 (about the same as before the crisis) to 7 (much worse than before the crisis). Relative effect of the crisis is also a scale: ‘Many people in Iceland have had financial problems because of the crisis. What effect do you believe the crisis has had on your financial situation compared to other Icelanders?’ The scale ranges from 1 (much better effect) through 5 (same effect) to 11 (much worse effect). Finally, we measure negative expectations using the mean for two questions: (1) ‘Do you believe that your standard of living will in the near future …’ and (2) ‘Do you believe that your status in society will in the near future …’. Responses range from 1 (get much better) to 5 (get much worse). The combined index ranges from 1 (highly positive expectations) to 5 (highly negative expectations). Cronbach’s alpha is 0. 070.

Control Variables

Household income is measured with a ranked scale indicating total monthly household income, ranging from 1 (lowest income category) to 17 (highest income category). We use four dichotomous variables for occupational status: part-time employment, currently unemployed, currently receiving pension or disability, and other status (student, homemaker, other), with full-time employment as the baseline group. We measure educational attainment using two dichotomous variables, secondary education and university education, with primary education (or less) as the baseline group. Gender is coded 1 for women and 0 for men. We allow for a nonlinear effect of age by including both age and age squared in the regression analysis. A dichotomous variable is coded 1 for individuals living in the Reykjavík metropolitan area, and 0 if otherwise. We measure number of children on a scale ranging from 0 (no child in household) to 6 (six or more children in household). Finally, we use a dichotomous variable for marital status, coded 1 for those who are married or living with a partner, and 0 otherwise. Table 1 shows descriptive statistics for all variables.

Table 1.

Descriptive statistics (N = 812).

	Mean	Standard deviation	Minimum value	Maximum value
Negative subjective comparisons
Change in standard of living	4.1	1.1	1	7
Relative effect of the crisis	5.5	1.9	1	11
Expectations about future outcomes	3.1	.66	1	5
Social demographics
Sex (Female = 1)	.51	.50	0	1
Age	45.5	16.6	18	92
Marital status (married/cohabiting = 1)	.53	–	0	1
Number of children in household	.85	1.1	0	6
Primary education (baseline group)	.32	–
Educational attainment
Primary education (baseline group)	–	–	–	–
Secondary education	.38	–	0	1
University education	.30	–	0	1
Capital area	.65	–	0	1
Household income	10.2	4.5	1	17
Employment status
Full-time work (baseline group)	–	–	–	–
Part-time work	.14	–	0	1
Unemployed	.03	–	0	1
Receives pension/disability	.10	–	0	1
Other status	.12	–	0	1
Dependent variables
Subjective injustice	2.6	.96	1	5
Anger	1.4	.36	1	3.40
Depression	1.7	.71	1	4

Results

The three comparison processes exhibit different distributions in our data. A majority of the respondents (62%) experienced a reduction in their standard of living during the crisis (Figure 2), while only 18 percent believe that the crisis has had a worse effect on them than on others (Figure 3). The distribution of respondents’ expectations of the future is slightly more balanced; 35 percent anticipate a negative change in their standard of living in the near future, and 11 percent think that their social status will worsen (Figure 4). That these comparison processes are so differently distributed may have important implications if indeed they turn out to interact in their effects on distress. For example, if social comparisons condition the effect of perceived reduction in living standard on distress, the effect of perceived reduction on distress may be quite modest because so few individuals experience negative social comparisons in this setting. We turn to this issue again after testing the hypotheses.

Figure 2.

Distribution of perceived change in standard of living during the crisis.

Figure 3.

Distribution of perceived relative impact of the crisis.

Figure 4.

Distribution of expectations about future outcomes.

We test the hypotheses presented in Figure 1 using regression models presented in Table 2. We estimated three models, one for each dependent variable: subjective injustice, anger, and depression. All models include the main effects of the three comparison processes (which have been grand-mean centered), the multiplicative terms for the predicted interaction effects, and the control variables.⁵

Table 2.

Least squares regression models for the effects of subjective comparisons on the indicators of distress.

	Dependent variables
	Subjective injustice		Anger		Depression
	Model 1		Model 2		Model 3
Independent variables	Regression coefficients	Standardized coefficients ^b	Regression coefficients	Standardized coefficients ^b	Regression coefficients	Standardized coefficients ^b
Negative subjective comparisons
Negative change in standard of living^a	.14***	.16	.02	.05	.08**	.13
Relative effect of the crisis^a	.03	.05	−.01	−.04	−.00	−.01
Negative expectations^a	.19***	.13	.03	.07	.02	.02
Interaction terms
Negative change in standard of living TIMES Relative effect of the crisis	.06***	–	.02***	–	.02	–
Negative change in standard of living TIMES Negative expectations	.08*	–	.03**	–	.05	–
Control Variables
Sex	.06	–	.01	–	.11*	–
Age	.05***	–	−.01	–	.00	–
Age squared	−.001***	–	.00	–	−.00	–
Married/relationship	.03	–	.01	–	−.08	–
Children	.02	.02	.03**	.10	.01	.01
Secondary education	−.24**	–	−.02	–	−.11	–
University education	−.31***	–	−.04	–	−.15*	–
Capital area	.11	–	.07	–	.01	–
Household income	−.03***	−.16	−.01*	−.09	−.02**	−.11
Part–time work	.07	–	.00	–	−.07	–
Unemployed	.10	–	.16*	–	.06	–
Receives pension/disability	−.08	–	−.03	–	−.03	–
Other status	.01	–	.01	–	.08	–
Constant	2.0		1.6		2.0
R²	.17		.14		.07

Notes: ^aMeasures for subjective comparisons are grand-mean centered and can be interpreted as main effects. ^bStandardized coefficients are reported only where appropriate; that is, for continuous independent variables only. *p< .05; **p< .01 (two-tailed tests).

Model 1 shows the effects of the three subjective comparisons on subjective injustice. The results lend some support for the hypotheses. As predicted (Figure 1, path a), perceived reduction in living standards has a significant and modest main effect on subjective injustice (beta = 0.16). Also as predicted (Figure 1, path e), negative expectations about future outcomes have significant main effects on subjective injustice (beta = 0.13). However, perceived relative effect of the crisis has no significant effect on subjective injustice, contrary to prediction (Figure 1, path d).

Both the interaction terms are statistically significant. The effect of perceived reduction in living standards on subjective injustice interacts with the other two comparison processes. Thus as predicted (Figure 1, path b), the effect of perceived reduction in living standards on subjective injustice is significantly more pronounced when individuals believe they have been harmed more by the crisis than others. Also as predicted (Figure 1, path c), perceived reduction in living standards has significantly more pronounced effect on subjective injustice as individuals have more negative expectations. Finally, the control variables influence subjective injustice, as we would have predicted. There is significantly less subjective injustice among the educated and among those having higher incomes. Finally, age has a nonlinear effect on subjective injustice; subjective injustice increases up to about age 50 and then decreases with age.

We further examine these interaction effects in Figures 5 and 6. Figure 5 displays the effect of perceived change in living standards on subjective injustice in three subgroups: individuals who feel the crisis had a better effect on their situation than others, those who feel it had a similar effect on them, and those who feel it has had a worse effect on their situation than others. The results illustrate the pattern of interaction effects. A perceived drop in living standards has the most pronounced effect on subjective injustice in the group of people who think they have been harmed more by the crisis than others (beta = 0.27). The effect is slightly weaker among those who feel the crisis has affected them similarly to others (beta = 0.21). No significant effect emerges among those who think that the crisis has harmed them less than others.

Figure 5.

The partial slope effect of perceived change in standard of living on subjective injustice in three subgroups: those who think that the crisis has harmed them less than most others (N = 292), similar to most others (N = 375), and more than most others (N = 145). All control variables from Table 2 have been set at their grand-means.

Figure 6.

The partial slope effect of negative change in standard of living on subjective injustice in three subgroups: those who expect positive change (N = 179), those who expect no change (N = 338), and those who expect negative change (N = 295). All control variables from Table 2 have been set at their grand-means.

Figure 6 shows the effect of perceived reduction in standard of living on subjective injustice in three subgroups: those who expect a positive change in their economic status in the near future, those who expect no change, and finally those who expect a negative change in their economic situation. The results show that perceived drop in living standard is moderately associated with subjective injustice in the group of individuals who expect their situation to worsen in the near future (beta = 0.27), but no such effects are observed in the other groups.

We now turn to Model 2 in Table 2, which presents the regression of anger. Although the comparison processes have no significant main effects on anger, there are significant interaction effects. Thus, perceived reduction in living standards has a significantly more pronounced effect on anger if individuals 1) think that the crisis has harmed them more than others; and 2) have negative expectations about future outcomes. Results for the control variables indicate that having children is associated with increased anger as is low household income and unemployment.

Again, we further examine the interaction effects by testing the effect of perceived change in living standard on anger in subgroups. Figure 7 shows how the effect of perceived negative change in living standards on anger is contingent on the perception of the relative impact of the crisis. This effect is significant and moderately strong for those who think that the crisis had a worse effect on them than on others (beta = 0.26). By contrast, a perceived drop in living standards has no significant effect on anger in the other two groups. Figure 8 shows that a perceived drop in living standards influences anger only among those who believe that their economic situation will worsen in the near future (beta = 0.18). These results further support the pattern of interaction effects that we have proposed above (again, see Figure 1, paths b and c).

Figure 7.

The partial slope effect of negative change in standard of living on anger in three subgroups: those who think that the crisis has harmed them less than most others (N = 293), similar to most others (N = 376), and more than most others (N = 147). All control variables from Table 2 have been set at their grand-means.

Figure 8.

The partial slope effect of negative change in standard of living on anger in three subgroups: those who expect positive change (N = 179), those who expect no change (N = 340), and those who expect negative change (N = 297). All control variables from Table 2 have been set at their grand-means.

Finally, the results for depression lend only partial support for our hypotheses. Thus Model 3 indicates a modest main effect of a perceived reduction in the standards of living on depression (beta = 0.13). The other comparison processes do not have significant effects on depression. Moreover, the interaction terms are not significant; the association between a perceived reduction in living standards and depression is not significantly contingent on the other comparison processes. Finally, we find significantly less depression among the educated and among high income earners.

The Buffering Effects of Subjective Comparisons

The results in Figures 5 -8 illustrate how the two processes, comparison to others and expectations, buffer the overall (main) effect of perceived reduction on subjective injustice and anger. Consider the moderating effect of social comparisons shown in Figures 5 and 7, while perceived reduction has the most pronounced effect in the group who thinks that the crisis has harmed them more than most others, only 18 percent of the respondents belong to that group, and thus the overall (main) effect of perceived reduction is reflected more by the weak effect observed among those who experience a more favourable social comparison. Likewise, in Figures 6 and 8, a perceived reduction in standard of living influences subjective injustice and anger only in the group which has negative expectations, reducing the overall (main) effect of perceived reduction on these indicators of distress.

Conclusion

Inspiring our study is the classic idea that sudden social changes, economic crises in particular, create distress as a result, not only of objective economic deprivation, but also of subjective deprivation (Davies, 1962; Durkheim, 1951[1897]). We have extended this idea by building on relative deprivation theory, arguing that economic crises can evoke subjective comparisons that condition the effects of suddenly worsening conditions on distress. We have used the economic collapse in Iceland as a test case for our argument, examining how three types of subjective comparisons influenced subjective injustice and emotional distress during the midst of the recession in Iceland.

The analysis reveals two key patterns. First, we find partial support for the idea that comparison to the situation of others conditions the effect of perceived reduction in the standard of living on distress. Thus, we find that a perceived reduction in the standard of living has only modest overall effects on subjective injustice and depression, and no overall effect on anger. However, perceived reduction in standard of living has a more pronounced effect on subjective injustice and anger among individuals who think that the crisis has harmed them more than others. In fact, a perceived reduction in standard of living has no effects on subjective injustice and anger if individuals believe that the crisis has affected them less than others. However, we find no interaction effects for depression. A perceived reduction in standard of living is the only comparison process significantly influencing depression in our data.

Second, expectations about future outcomes moderate the effects that a perceived reduction in the standard of living has on subjective injustice and anger. A perceived reduction in the standard of living influences subjective injustice and anger more if individuals have negative expectations, but it has no effect on these indicators of distress among those who have positive expectations. Again, we find no interaction effects for depression.

Our study contributes to social research in a few respects. It contributes to research focusing on the link between economic crises and mental health (Brenner, 1987, 1990; Davalos and French, 2011; Dooley et al., 1994; Durkheim, 1951[1897]; Tausig and Fenwick, 1999; Zivin et al., 2011). We are among the first to examine the role of subjective comparisons in linking crises and distress (see Krahn and Harrison, 1992). Our findings suggest that social comparisons and expectations may buffer the effect of worsening living conditions on individual distress. Thus when there is a widespread perception that most people have been hit hard by a crisis, most individuals may tend to accept their economic loss because they think that others have been hit at least just as badly. A large majority of Icelanders believe that the crisis has harmed them similarly or less than it has harmed others, apparently producing a weak overall effect of perceived reduction in standard of living on subjective injustice and anger. Ironically, a widespread perception of despair creates a social context where even a large drop in the standard of living has only a small overall effect on individual distress. Likewise, the effects of worsening living conditions on distress seem contingent on whether the social context encourages pessimism about the future.

The findings may provide a clue as to why economic crises sometimes have only weak effects on mental health (e.g. see Davalos and French, 2011). The effects of crises on well-being may depend on social contextual factors influencing 1) how many individuals think they have been hit worse by the crisis than others; and 2) how many individuals are pessimistic about their status in the near future. Crises may increase distress more when these perceptions are widespread. Such perceptions in turn may be shaped by various factors, including media coverage and political reactions, which may or may not reflect the objective reality of a particular crisis. Our individual-level study cannot address these issues directly, and thus more research is needed.

Our study demonstrates how relative deprivation theory helps to link broad social conditions with individual well-being, contributing to scant direct research on the issue. Although social comparisons are held to play a role in mediating macro social conditions, including inequality (Bernburg, 2010; Blau and Blau, 1982; Wilkinson and Pickett, 2007; Yngwe et al., 2005) and economic crises (Davies, 1962), on distress, research on this issue rarely uses direct measures of subjective comparisons. Our use of direct measurement suggests that social comparisons can sometimes buffer the effect of disadvantage on distress. Again, the widespread perception that one’s loss during the crisis is no worse than the loss of others appears to buffer the overall effect of perceived economic loss on some forms of distress in the Icelandic crisis.

This study has limitations. First, it is non-experimental and does not directly test causal statements. We can examine only whether the statistical patterns are consistent with our theoretical argument. Moreover, relying on cross-sectional data has prevented us from measuring the focal constructs in a temporal order. Although our model focuses on here-and-now perception of current conditions in relation to three different reference points, measuring subjective injustice and distress before as well as during crisis would provide a stronger test of our model. Moreover, although we control for the current economic situation, assuming that perception of change in the standard of living is largely reflective of objective change in the standard of living (Krahn and Hartnagel, 1992), ideally we need to include objective economic change in the model.

It remains to be seen whether subjective comparisons constitute ‘fundamental mechanisms’ (Hedström and Swedberg, 1998) that generally help to explain how sudden change influences distress and well-being. We acknowledge that other comparison processes may also play a role, including, for example, cross-national comparisons (Whelan et al., 2001); that is, comparisons with citizens of other countries less affected by the global crisis. In conclusion, examining the interplay of subjective comparisons may help future research to understand how individuals experience sudden social change. Iceland has been an ideal setting to examine this issue. The Icelandic crisis was severe and occurred suddenly after a long period of rising expectations, and it was related to images of corruption in public opinion. This situation often characterizes major economic crises. We call on future research to apply our model in settings characterized by economic crises and other types of abrupt changes.

Footnotes

Acknowledgements

We thank Clayton Fordahl, Janet Gornick, Patrick Sachweh, Jason Schnittker, Heather Smith, Thorolfur Thorlindsson, and two anonymous reviewers for helpful comments. An earlier version of this paper was presented at the XVII World Congress of Sociology in Goteborg Sweden in July 2010.

Funding

The study was aided by grants from the Icelandic Centre for Research, Edda – Center of Excellence, and University of Iceland Research Fund.

Notes

Berglind Hólm Ragnarsdóttir is a doctoral student in sociology at the City University of New York, Graduate Center. Her research interests include social stratification and inequality, social psychology and gender. She is currently a research associate at LIS Cross-National Data Center where her work focuses on projects related to cross-national stratification, inequality and gender and work.

Jón Gunnar Bernburg is a professor of sociology at the University of Iceland. He has a PhD in sociology from the State University of New York at Albany. He has studied topics in several areas of sociology, including labelling and stigma, anomie, relative deprivation, and neighborhood effects. His current work focuses on the economic crisis in Iceland. Jon Gunnar has published in various journals, including Social Forces, Social Science and Medicine, and Criminology.

Sigrun Olafsdottir is an assistant professor of sociology at Boston University. Her research stands at the intersection of medical, political, and cultural sociology. She is particularly interested in how larger political and cultural contexts influence individual outcomes and attitudes. She is the Principal Investigator on an NIH-funded grant, titled ‘A Comparative Approach to Health Inequalities’. Her work has appeared in numerous journals, including Journal of Health and Social Behavior, Social Science Medicine, and European Sociological Review.

References

Benediktsdóttir

Daníelsson

Zoega

(2011) Lessons from a collapse of a financial system. Economic Policy 26(66): 183–235.

Bernburg

(2010) Relative deprivation theory does not imply a contextual effect of country-level inequality on poor health: A commentary on Jen, Jones, and Johnston (68:4, 2009). Social Science & Medicine 70: 493–95.

Bernburg

Thorlindsson

Sigfusdottir

(2009) Relative deprivation and adolescent outcomes in Iceland: A multilevel test. Social Forces 87: 1223–50.

Blau

(1982) The cost of inequality: Metropolitan structure and violent crime. American Sociological Review 47: 114–29.

Brenner

(1987) Relation of economic change to Swedish health and social well-being, 1950–1980. Social Science & Medicine 25: 183–95.

Brenner

(1990) Influence of the economy on mental health and psychophysiological illness: International perspective. Community Mental Health in New Zealand 5: 2–10.

Burgard

Brand

House

(2009) Perceived job insecurity and worker health in the United States. Social Science & Medicine 69: 777–85.

Crosby

(1976) A model of egoistical relative deprivation. Psychological Review 83: 85–113.

Davalos

French

(2011) This recession is wearing me out! Health-related quality of life and economic downturns. Journal of Mental Health Policy and Economics 14: 61–72.

10.

Davies

(1962) Toward a theory of revolution. American Sociological Review 27: 5–19.

11.

Dooley

Catalano

Wilson

(1994) Depression and unemployment: Panel findings from the epidemiologic catchment area study. American Journal of Community Psychology 22: 745–65.

12.

Durkheim

(1951[1897]) Suicide: A Study in Sociology. New York: Free Press.

13.

Folger

(1984) Perceived injustice, referent cognitions and the concept of comparison level. Representative Research in Social Psychology 14: 88–108.

14.

Gartrell

(2002) The embeddedness of social comparison. In: Walker

Smith

(eds) Relative Deprivation. Specification, Development and Integration. Cambridge: Cambridge University Press, 164–84.

15.

Graziano

(1987) Lost in thought at the choice point: Cognition, context, and equity. In: Masters

Smith

(eds) Social Comparison, Social Justice, and Relative Deprivation. London: Lawrence Erlbaum, 265–94.

16.

Hedström

Swedberg

(1998) Social mechanisms: An introductory essay. In: Hedström

Swedberg

(eds) Social Mechanisms: An Analytical Approach to Social Theory. Cambridge: Cambridge University Press, 1–31.

17.

Jónsson

(2011) Boom and Crisis in Iceland through the Prism of Consumption. Presented at the Crisis and Renewal conference, University of Iceland, 2–3 June November.

18.

Kish

Frankel

(1974) Inference from complex samples. Journal of the Royal Statistical Society 36: 1–37.

19.

Krahn

Harrison

(1992) ‘Self-referenced’ relative deprivation and economic beliefs: The effects of the recession in Alberta. Canadian Review of Sociology 29: 191–209.

20.

Levine

Moreland

(1987) Social comparison and outcome evaluation in group context. In: Masters

Smith

(eds) Social Comparison, Social Justice, and Relative Deprivation. London: Lawrence Erlbaum.

21.

Lind

Tyler

(1988) The Social Psychology of Procedural Justice. New York: Plenum Press.

22.

Markovsky

(1985) Toward a multilevel distributive justice theory. American Sociological Review 50: 822–39.

23.

Matthíasson

(2008) Spinning out of control: Iceland in crisis. Nordic Journal of Political Economy 34: 1–19.

24.

Merton

(1968) Social structure and anomie. In: Merton

(ed.) Social Theory and Social Structure. New York: Free Press, 185–214.

25.

Merton

Rossi

(1968) Contributions to the theory of reference group behavior. In: Merton

(ed.) Social Theory and Social Structure. New York: Free Press, 279–440.

26.

Mirowsky

(1987) The psycho-economics of feeling underpaid: Distributive justice and the earnings of husbands and wives. American Journal of Sociology 92: 1404–34.

27.

Reis

(1987) The nature of the justice motive: Some thoughts on operation, internalization, and justification. In: Masters

Smith

(eds) Social Comparison, Social Justice, and Relative Deprivation. London: Lawrence Erlbaum.

28.

Report of the Special Investigation Commission (2010) The Causes of the Collapse of the Icelandic Banks 2008 and Related Events. Reykjavík: Special Investigation Commission.

29.

Rosenfield

Smith

(2010) Gender and mental health: Do men and women have different amounts or types of problems? In: Scheid

Brown

(eds) A Handbook for the Study of Mental Health: Social Contexts, Theories, and Systems. Cambridge: Cambridge University Press, 256–67.

30.

Runciman

(1966) Relative Deprivation and Social Justice. London: Routledge and Kegan Paul.

31.

Skitka

(2009) Exploring the ‘lost and found’ of justice theory and research. Social Justice Research 22: 98–116.

32.

Smith

Ortiz

(2002) Is it just me? The different consequences of personal and group relative deprivation. In: Walker

Smith

(eds) Relative Deprivation. Specification, Development and Integration. Cambridge: Cambridge University Press, 91–118.

33.

Smith

Pettigrew

Pippin

Bialosiewicz

(2011) Relative deprivation: A theoretical and meta-analytic review. Personality and Social Psychology Review. doi: 10.1177/1088868311430825

34.

Statistics Iceland (nd.a) An Overview of Gross Domestic Product and Gross National Product from 1945-2011. Available at: http://www.statice.is

35.

Statistics Iceland (nd.b) Employment, Unemployment and Labour Force – Original Data – Monthly 2003–2012. Available at: http://www.statice.is

36.

Stouffer

Schuman

DeVinney

Star

Williams

(1949) The American Soldier: Adjustment During Army Life, vol.1. Princeton, NJ: Princeton University Press.

37.

Tajfel

(1978) Differentiation between Social Groups: Studies in the Social Psychology of Intergroup Relations. London: Academic Press.

38.

Tausig

Fenwick

(1999) Recession and well-being. Journal of Health and Social Behavior 40: 1–16.

39.

Testa

Major

(1990) The impact of social comparisons after failure: The moderating effects of perceived control. Basic and Applied Social Psychology 11(2): 205–18.

40.

United Nations Development Programme (2007) Human Development Report 2007/2008. New York: Palgrave Macmillan.

41.

Yngwe

Fritzell

Burström

Lundberg

(2005) Comparison or consumption? Distinguishing between different effects of income on health in Nordic welfare states. Social Science and Medicine 61: 627–35.

42.

Whelan

Layte

Maitre

Nolan

(2001) Income, deprivation, and economic strain. An analysis of the European community household panel. European Sociological Review 17: 357–72.

43.

Wilkinson

Pickett

(2007) The problems of relative deprivation: Why some societies do better than others. Social Science & Medicine 65: 1965–78.

44.

Zivin

Paczkowski

Galea

(2011) Economic downturns and population mental health: Research findings, gaps, challenges and priorities. Psychological Medicine 41: 1343–8.