1Welfare economics or welfarism is an extension of classic utilitarianism. But, whereas according to classic utilitarianism (Bentham, 1823 [1789]), optimal social well-being is calculated on the basis of maximizing pleasure and minimizing pain, welfarism, according to Amartya Sen’s definition (1999, 59), restricts “the judgments of state of affairs to the utilities in the respective states.” In short, the concept of well-being here incorporates activities that may not bring us any pleasure, nor any pain. The key is to take account of our preferences, even when they concern society (for example, if we would like a more egalitarian society) and not just our own situation. Social or collective well-being does not simply maximize the sum of individual utilities or interests, but it satisfies, as far as possible, the preferences of the greatest number. How, though, are individual preferences to be aggregated to achieve collective well-being? How, to this end, are these preferences to be weighed against one another? How should costly preferences be dealt with? How are preferences to be ranked without making paternalistic judgments about them? These questions are left unresolved by welfarism and continue to be the subject of debate. Moreover, Rawls’ well-known objection (2005 [1971], 3–4) to utilitarianism seems equally valid for its extension in welfarism:

2Each person possesses an inviolability founded on justice that even the welfare of society as a whole cannot override. For this reason, justice […] does not allow that the sacrifices imposed on a few are outweighed by the larger sum of advantages enjoyed by many. […] the rights secured by justice are not subject to political bargaining or to the calculus of social interests.

3This is why for Rawls and those who follow him to some extent (Dworkin 1981; Pogge 2008 [2002]), if well-being is to be maximized, it must be understood as the maximization of the resources (the origin of the name “resourcism” given to this school of thought) of the most disadvantaged. Sen’s response (2009) is different. According to him, well-being is the realization of “capabilities” or “the power to be” and “to do,” which involves taking account of the real freedom individuals have to live the life they have chosen according to their preferences. At the level of society as a whole, this has led to criticism of measuring well-being using GDP alone. Other dimensions should be taken into account. The Human Development Index (HDI), calculated by the United Nations Development Programme (UNDP) under the guidance of Sen (among others), therefore also takes into account health (life expectancy) and level of education. These three factors (GDP, health, and education), weighted to produce a single index, nevertheless remain insufficient. Moreover, UNDP reports are, in fact, significantly more comprehensive in their analyses. The Stiglitz-Sen-Fitoussi report (2009) goes even further in its recommendations, proposing in particular that the measurement of subjective well-being should not be overlooked. At the individual level, measures have, in fact, often been based on “objective” well-being, using a range of indicators that ultimately consider resources, such as employment, participation in political life, and income. But this limitation turns out to be contentious, in particular from the perspective of capabilities, as it seems equally imperative to consider how individuals themselves judge their own well-being. This time, the measure is not based on a list of items chosen a priori as necessary for well-being, but it is found a posteriori by analyzing the responses to questions concerning the subjective assessment of well-being posed to representative samples. This ultimately leads to a highly multi-dimensional conception of well-being (individual, collective, subjective, objective).

4Does this, in fact, answer Rawls’ objection? That is far from certain. However, it is not from a theoretical perspective that we seek to approach this debate. By keeping to the subjective aspect, which must be taken into account, we will limit ourselves to considering, empirically, what the relationship might be between subjective well-being and feelings of social justice. Our observations will not therefore provide all the answers to the question of how to understand the interconnection between well-being and justice. They will, however, clarify the perspective of the stakeholders, a dimension that cannot be overlooked.

Two approaches to satisfaction

5Among the wealth of research on subjective well-being, we can identify two broad approaches. On the one hand, we find studies focused on feelings and emotions. These involve almost “objectively” measuring the positive and negative moments experienced during the course of a day, by counting, for example, smiles or time spent on enjoyable or unpleasant activities, by identifying periods of depressed mood, or by analyzing sleep quality. In this vein, Daniel Kahneman and Alan Krueger (2006) have proposed an index based around unhappiness, by measuring the proportion of time an individual spends in an unpleasant state. On the other hand, we have studies that focus on individuals’ own assessment of their overall level of life satisfaction. Individuals are asked for their judgment about their own quality of life.

6Although these two approaches are not without their similarities, each has its own specificities. The first prides itself on being an objective measure, but in practice it has proven to be particularly sensitive to the problem of adaptation to circumstances: everyone adjusts their expectations to match what they think they can actually achieve; and, however high their expectations, everyone will have their share of ups and downs. The most fortunate will not smile at the same things as the less fortunate, nor will they lament the same misfortunes. Yet—pushing the adaptation argument to its limit—the emotional balance sheet could be the same. Furthermore, such an approach focuses exclusively on the present moment and leaves a lifetime of experiences in the shadows. The second approach is clearly presented as a subjective measure and, therefore, the problem of adaptation seems less pronounced (Deaton 2008). Each person judges for themselves, of course, but this judgment involves making comparisons and having a rationale for being satisfied or dissatisfied with life. This type of measure is therefore not purely subjective. Despite all this, judging one’s own well-being is not an easy task and, ultimately, it would have to be possible to take into account all aspects of life. It is therefore unsurprising, as observed by several writers (Deaton 2011; Schwarz and Strack 1999), that the way in which the question is asked influences the response. There is a framing effect: respondents judge their quality of life taking into account the aspects of life that the interviewer has previously brought to mind in their questioning.

7We will nevertheless follow this second approach here. However, in order to make an empirical connection between subjective well-being measured in this way and feelings of social justice, we are confronted with a difficulty; very few statistically representative surveys pose these two questions together. The “Perception des Inégalités et les Sentiments de Justice” (PISJ) [Perception of Inequality and Feelings of Justice] survey is an exception in this regard and will be our primary source for our analysis (Forsé et al. 2013). It was conducted at the end of 2009 by the market research company GfK-ISL, with a representative sample of 1,711 people aged eighteen or over living in France. The question on well-being was highly standard in its formulation, in that it reproduced the wording found in many other surveys, in particular international surveys: “All things considered, how satisfied or dissatisfied are you with your life these days? Use a scale from 1 to 10, where 1 is ‘not at all satisfied’ and 10 is ‘completely satisfied.’”

Subjective well-being in France and Europe

8Given that the average response gave a score of 6.4, we can assert that French people are on the whole quite satisfied with their lives. But is this result unique to France?

9The same question was posed in the European Values Survey (EVS) in 2008 (or 2009 in some countries). The level of satisfaction in France (7.1 in 2008) is a little higher than the level found in the PISJ survey in 2009. It should however be noted that the question is asked nearer the start of the EVS questionnaire, whereas it is at the end of the PISJ survey—suggesting that some kind of reassessment takes place after numerous inequalities have been brought to mind. Furthermore, the effects of the subprime mortgage crisis were very likely felt to a greater degree in 2009 than in 2008, and it is possible that this led to a higher level of dissatisfaction. In any event, the French well-being score according to the EVS is equal to the average (7.1) for the rest of the thirty-six European countries where the survey was conducted. In terms of well-being, France therefore seems to be an average European country.

10Ranking the average satisfaction scores in these countries enables us to loosely divide them into three broad groups. In Northern Europe, perceived well-being is clearly higher than the average. In Southern Europe (including France), it is nearer the average, and in Eastern European countries (as well as in Germany), the respondents stated that they are less satisfied with life. These groups are of course not perfectly homogenous. Some countries are exceptions and have a level of well-being that puts them in a group to which they would not belong geographically. But these exceptions are quite rare and the general trend is as described. Incidentally, the highest level of satisfaction was recorded in Denmark and the lowest in Bulgaria.

11Another survey conducted in 2012, the Gallup World Poll, shows that Northern European countries continue to have the highest levels of satisfaction. In France, the level of satisfaction is equal to the average for OECD countries. Germany has risen to the level of France and the United Kingdom, at around this average. The highest levels of dissatisfaction are still recorded in countries in Eastern Europe, but they are now joined by the Southern European countries (Italy, Spain, Greece, and Portugal). This sharp fall in well-being between 2008 and 2012 is undoubtedly a consequence of the economic crisis that, as is well known, hit these countries hard, leading to a significant loss of purchasing power and high unemployment.

Three dimensions of satisfaction

12By studying satisfaction in French society in greater detail, we find that it clearly depends on a large number of factors (Manzo 2011). We note, as a starting point, certain demographic and geographical elements. There is no difference between the sexes. On average, men are as satisfied with their lives as women (6.4 and 6.3 respectively, a gap within the margin of error). In contrast, average satisfaction varies slightly by age group. Satisfaction gradually falls with age until people reach their fifties, or thereabouts, before rising again. As such, average satisfaction reaches 6.4 for 18–24 year olds, falls to 5.7 for 48–53 year olds, and then rises to 6.8 for 60–65 year olds and 6.6 beyond that. This U curve has already been observed in other data (Blanchflower and Oswald 2007). Average satisfaction also varies slightly depending on the size of the residential area in which a person lives: people who live in major cities (6.2) or their suburbs (6.2) appear a little less satisfied than other people, in particular those who live in rural areas (6.6).

13In order to understand these variations, the determinants of satisfaction must be considered more closely. They come, you could say, in all shapes and sizes. We began by grouping them into three dimensions. The first concerns individual resources, both in financial terms and in terms of physical and mental health (however, the health aspect is beyond the scope of our study and will not be discussed further in this article). The second concerns employment, quality of working life, and feelings of success and recognition in relation to one’s work and one’s career trajectory. Finally, the last dimension groups together various aspects of people’s emotional and social lives: partners, children, friends, and participation in clubs or other groups.

The importance of standard of living

14Each of these dimensions is a field of research in its own right. Accordingly, the influence of income on well-being has been explored extensively, both at the individual level and the societal level, ever since Richard Easterlin’s renowned article (1974) which, using data from a range of sources, concluded that economic growth does not make the population more satisfied. We are, of course, referring to the Easterlin Paradox. Since then, with data having been collected in almost every country, we can no longer be in any doubt about the major role that income plays in life satisfaction. Not only are the richest societies also those in which the population is on average the most satisfied with life, but within each society, the richest individuals are more satisfied with life than the poorest. The Easterlin Paradox asserts more specifically that the influence of income on satisfaction is much weaker at the societal level than at the individual level within a society. From this, we could conclude that the benefit of economic growth seems to be lost, because life satisfaction seems in essence to be relative to the situation of one’s fellow citizens. But this hypothesis is not confirmed by the most recent work on this subject. It seems, rather, that the influence of income on satisfaction may lie between the individual level and the societal level (Hagerty and Veenhoven 2003; Deaton 2008; Stevenson and Wolfers 2008).

15However we analyze it, there is no shadow of a doubt about the influence of income or standard of living on satisfaction. According to the PISJ survey, the correlation between the various indicators of income or standard of living and the level of satisfaction is always very strong. There is simply no other variable more correlated with satisfaction than these indicators. As such, the (Pearson) correlation between satisfaction and standard of living (income per consumption unit) is 0.31. The correlation even increases to 0.35 when we use the standard of living logarithm. Furthermore, all the studies tend to demonstrate that logarithmic scales should be used for income or standard of living where the relationship with satisfaction is concerned. The satisfaction obtained from money seems in fact to be more relative than absolute: a rich person would not find a gift of one hundred dollars as satisfying as a poor person would; whereas both the rich person and the poor person would be approximately equally satisfied with a gift of 10% of their respective incomes.

16The more subjective indicators used to estimate standard of living are also closely related to life satisfaction. The correlation between satisfaction and experiencing budget difficulties at the end of the month is very strong (0.40). Likewise, there is a strong correlation between satisfaction and the feeling of being disadvantaged in terms of income inequality (0.42), wealth inequality (0.30), and even housing inequality (0.31). The correlation with standard of living (on a logarithmic scale) is 0.35. The respondents were also asked to estimate in which personal income quartile they belong. Given that they are not often wrong about their position, the correlation between this self-ranking and satisfaction is also high (0.30). Table 1 broadly illustrates the differences in satisfaction level caused by income inequality. This difference is around 2 points between poor households and affluent households, which corresponds to the standard deviation for this measure of satisfaction.

17The influence of income on satisfaction has several aspects and is clearly not just a question of purchasing power and consumption. Beyond income, there are questions concerning the quality of employment, recognition, and trust, which influence satisfaction in their own way. Nevertheless, the purely monetary aspect certainly cannot be disregarded. People need financial resources to be able to live life as they want, and the greater access they have to these resources, the more likely they are to be satisfied. Experiencing budget difficulties at the end of the month is, moreover, one of the most significant determinants of dissatisfaction. A lack of monetary resources also feeds into dissatisfaction in relation to the feelings of deprivation experienced when confronted with all the new products that consumer society endlessly offers up. A question about deprivation in relation to an endless supply of new products was asked in the PISJ survey and it was observed that there is a correlation between life dissatisfaction and this feeling that the growth in new products is a source of deprivation (0.25).

18These results demonstrate that the adaptation of preferences to life circumstances does not go so far as to entirely erase the difference in satisfaction between privileged people and disadvantaged people. The adaptation argument, which is generally associated with Sen’s work, is not however invalidated, rather, it has a limited scope. Individuals maintain a critical perspective on their lives, conceivably by comparing themselves with others, as we will see.

The impact of career progression and comparisons with others

19We now turn to the second dimension of subjective well-being described above. Life satisfaction is judged in relation to life goals, which are situated aspirations and functions of individuals’ initial levels of resources. This primarily concerns professional goals beyond the issue of salary (but salary should also be considered a form of recognition, without it being the only possible form). From this perspective, dissatisfaction increases throughout a person’s career. This can manifest itself in feeling drained, fatigued, or frustrated following successive failures or even “mediocre” success compared with initial aspirations. According to the PISJ, on average, individuals compare themselves slightly more with people who are more successful than themselves. While 60% of respondents stated that they have done just as well as their friends, 26% said they have done less well, and 13% said they have done better. Likewise, while 69% of respondents stated that they have done just as well as their colleagues or those in the same profession, 21% said they have done less well, and 10% said they have done better. Nevertheless, as we can see from Table 2, people’s tendency to compare themselves with people who are more successful grows throughout their career, perhaps simply because gaps in achievement become increasingly pronounced and visible with age.

Table 2

Responses to the question “As regards your personal income, do you think that you have done better than… 1) your friends… 5) people who are the same age as you” by respondent age

18-30 years 31-40 years 40-50 years 51-60 years >60 years Together Your friends Better 16 % 14 % 13 % 13 % 12 % 13 % Just as well 59 % 59 % 57 % 56 % 67 % 60 % Less well 25 % 27 % 31 % 31 % 22 % 27 % Total 100 % 100 % 100 % 100 % 100 % 100 % People who are the same age as you Better 24 % 18 % 16 % 14 % 20 % 19 % Just as well 53 % 55 % 52 % 56 % 65 % 57 % Less well 23 % 27 % 31 % 31 % 15 % 24 % Total 100 % 100 % 100 % 100 % 100 % 100 %

Responses to the question “As regards your personal income, do you think that you have done better than… 1) your friends… 5) people who are the same age as you” by respondent age

20Satisfaction depends rather strongly on these comparisons with others. The feeling of having done better, as regards personal income, than one’s friends is visibly gratifying (average satisfaction score of 7.1), a little more than having done just as well (6.7), but significantly more than having done less well (5.1). Of course, part of this gratification is linked to the fact of having a good income; nevertheless, even once experiencing difficulties at the end of the month has been controlled for, the correlation between satisfaction and relative position compared with friends remains quite strong (0.22, whereas the simple Pearson correlation is 0.32). If we now ask which reference groups are the most significant when individuals evaluate their own life satisfaction, we find first and foremost that friends are the main reference group, followed by individuals from the same generation (Pearson correlation of 0.31), colleagues and people in the same profession (0.26), neighbors (0.26), and, finally, parents (0.23).

21Consequently, life satisfaction is also linked to social status. Senior executives (6.9) and individuals in intermediate occupations (6.8) are more satisfied than workers and employees (5.9). Individuals’ opinions about their position in the social hierarchy is, in this regard, an even more influential factor. In the PISJ survey, we asked respondents to position themselves on a scale of social status, from 1 at the bottom of the hierarchy, to 10 at the top. This self-ranking is highly correlated with satisfaction (Pearson coefficient of 0.36). The correlation is also very strong with feelings of belonging to a class. People who stated that they are part of “the disadvantaged or marginalized classes” are clearly the most dissatisfied (3.8). They are followed by those who put themselves in the “working class” (5.6), then by those who identify as “lower-middle class” (6.3), “upper-middle class” (7.5), and, finally, “upper class, affluent” (7.6).

22It is not easy to disentangle the share of the effect of income linked directly to consumption and purchasing power from the share linked to status and recognition. But there can be little doubt that both have a strong impact on satisfaction. If we consider the question on experiencing budget difficulties at the end of the month as a good indicator of the “resources” dimension, and the question on belonging to a particular class as a good indicator of the “status” dimension, it seems then that these two dimensions are equally significant for subjective well-being: each has a Pearson correlation with satisfaction of 0.40. If we then calculate the partial correlation between satisfaction and each of its variables by controlling for the other variable, we get fairly similar results: the partial correlation between satisfaction and experiencing budget difficulties at the end of the month, controlling for the feeling of belonging to a class, is 0.29; and the partial correlation between satisfaction and the feeling of belonging to a class, controlling for experiencing budget difficulties at the end of the month, is 0.28. It therefore seems that each of these factors carries the same weight in terms of life satisfaction.

23Individuals’ feelings of justice as regards their own level of remuneration is also a determinant of satisfaction. People who stated that their remuneration is fair are more satisfied than others. Their average satisfaction score is 6.9, while those who said they are paid a little less than would be fair have a satisfaction score of 6.5, and those who said they are paid much less have a satisfaction score of 5.3. More surprisingly, but confirming other studies (Pritchard, Dunnette and Jorgenson 1972), people who stated that they are overpaid compared with what would be fair are not especially happy: their satisfaction score is 6.3. One possible explanation could be that these people feel that they are in an awkward position vis-à-vis their colleagues.

24Beyond salary, working conditions play a clear role. As such, for the question “in your working relationships, do you feel well treated by your direct superiors?”, those who said they feel poorly treated are clearly more dissatisfied than others (Pearson correlation of 0.26 [2]). Likewise, people who said they feel disadvantaged in terms of inequality regarding physical strain of work are more dissatisfied than others (Pearson correlation of 0.21).

25As would be expected, being unemployed is a very significant determinant of dissatisfaction (Pearson correlation of 0.34). The average satisfaction score of unemployed people is 5.3, whereas the average satisfaction score of employed people is 6.4. In the same vein, the satisfaction score of people who did not choose to work part-time is much lower (5.5) than that of those who chose to do so (7.1). Job security is a source of satisfaction, as demonstrated by the average difference in the satisfaction scores of permanent public-sector workers (6.8), public-sector interns or those in temporary roles or on leave (6.3), private-sector employees on permanent contracts (6.3), and private-sector employees on fixed-term contracts (6.1).

The roles of social mobility and relative deprivation

26Beyond an individual’s position in the social hierarchy, the feeling of having improved that position is also a source of subjective well-being. The survey asked the interviewees to estimate their current social status on a scale from 1 to 10, as well as their status ten years ago. The difference between these two scores provides a rough indicator for feelings of upward social mobility. This correlates fairly strongly with satisfaction (Pearson correlation of 0.25).

27We can also consider an indicator of social mobility that is a little more sophisticated by comparing, for each individual, absolute mobility—as calculated in this way—with the average mobility. This relative mobility can be assessed by decreasing a person’s current self-declared socio-economic status by that of ten years ago. This gives a least squares regression line that provides an estimate for the average mobility gradient. Individuals whose mobility positions them above this line can be considered to have experienced relative upward social mobility, whereas those whose mobility positions them below the line have not benefitted from such a degree of relative upward mobility. Technically, as Bernd Wegener (1991) proposes, this considers the residuals in the regressions as indicators of relative mobility.

28We therefore find that the correlation between relative mobility and subjective well-being (0.37) is much stronger than that of absolute mobility and subjective well-being. Of course, those with low relative mobility are more likely to be dissatisfied, whereas those with high mobility proved to be more satisfied. Given that individual mobility is compared here with overall mobility, this suggests that relative deprivation plays a significant role in the subjective evaluation of well-being.

29As such, one of the most conventional hypotheses of sociology is confirmed once again. Alexis de Tocqueville (1863 [1840], 166) had already clearly identified the underlying mechanism:

30When inequality of conditions is the common law of society, the most marked inequalities do not strike the eye: when everything is nearly on the same level, the slightest are marked enough to hurt it. Hence, the desire of equality always becomes more insatiable in proportion as equality is more complete.

31A survey conducted by Samuel Stouffer and his colleagues (1949) demonstrates, at least in part, the empirical validity of this reasoning. In their comparative study of two American army corps in the aftermath of the Second World War, they were confronted with surprising results. All the data made available to them indicated that members of the air force have much greater promotion prospects than members of the military police; yet, in the course of their interviews, some among the first group stated that they are very dissatisfied with their career opportunities, whereas those in the second group indicated that they are, to the contrary, satisfied. Stouffer explains this paradox by introducing the concept of the reference group. The air force personnel compared themselves with those who had been promoted, of which there are many in their branch, and thus formed a view of the standard career. Through this comparison, those who had not been promoted naturally experienced feelings of deprivation that led them to express dissatisfaction. The military police also compared themselves with the most common situation, but, on this occasion, this concerned people who had not been promoted. As such, those who had not been promoted did not experience any feelings of deprivation and did not express dissatisfaction.

32We can of course critique the psychological “proof” of Stouffer’s explanation by asking why the reference group is necessarily the largest group. Raymond Boudon (1977) asserts that it should rather be based on the objective prospect of promotion resulting from the structures of competition between individuals. When these structures leave open the possibility of promotion, according to the theory of expected utility, they lead individuals to invest in career advancement and actively participate in competition. However, there will necessarily be those who lose out, the proportion of which varies depending on certain properties of these structures. In this event, it is an investment made in vain that explains the feelings of deprivation. These individuals have been unable to achieve their objective as a result of the competition structure in which they find themselves and their satisfaction is consequently eroded.

33For his part, Garry Runciman (1966) does not fail to make the link with feelings of justice given that, even if certain structural conditions must be considered, it is fairly easy to understand that relative deprivation leads to a situation being considered unjust. But by establishing this relationship, he also seeks to demonstrate that this deprivation should not be understood in a solely “egoistic” sense, through comparison of one’s own position with that of a group. It is also “fraternalistic,” to the extent that the position of this group within society as a whole can also influence feelings of deprivation.

34In order to assess the impact of relative deprivation on subjective well-being, we will adopt our own understanding of this distinction, simply by changing the terminology to refer to “egocentric” relative deprivation for the former, and “social” relative deprivation, for the latter.

35One measure of egocentric relative deprivation can be obtained by comparing an individual’s salary with the salary that that individual claims is earned by those in the same profession, and considering the logarithm of this comparison. The correlation with subjective well-being is then 0.19. This score is not insignificant, but it is nonetheless clearly lower than that linking well-being with absolute or relative mobility. Is it therefore possible to find a stronger correlation if we turn to social relative deprivation? Here we can calculate the logarithm of the relationship between the salary that respondents think is paid in their profession and the salary that respondents would consider desirable for the same profession. In this case, the personal circumstances of the respondents are not directly at issue, rather the circumstances of people in the same profession. The correlation between this indicator and satisfaction then becomes weak at 0.07.

36Relative deprivation, whether egocentric or social, thus has a weaker influence on satisfaction than perceived social mobility over the course of the past ten years. Nevertheless, it is clear that this form of deprivation is not without effect. We have seen that comparisons with friends, colleagues, etc. are important, and it appears here that mobility compared with a reference—in this case, the average experience—has a greater impact than the same mobility considered in and of itself. There can therefore be no doubt that satisfaction is (also) a question of comparison. It is an individual’s perception of their own career advancement as more or less strong in relation to their reference groups that significantly impacts that individual’s feelings of well-being.

37It is interesting to note that relative deprivation is here much more egocentric than social, whereas when feelings of justice or injustice concern an individual’s own remuneration (i.e., feelings of microjustice), social deprivation plays a slightly more significant role. Using the same indicators as above, the correlation of feelings of microjustice with social relative deprivation is 0.16. This is overtaken by the correlation of microjustice with egocentric deprivation (0.21), with absolute mobility (0.18), and, even more so, with relative mobility (0.28). It is still the case that the impact of social relative deprivation on feelings of personal justice are stronger than on feelings of personal well-being. This is incidentally not illogical. As regards justice, how the reference group (or the group to which an individual belongs) is treated (here economically) by society can, as predicted by Runciman, have an effect on how that individual judges the fairness of their own remuneration. If I work in a profession that I judge to be underpaid on the whole, it is normal that I would consider myself to be unfairly paid. To take just one example, this is certainly a widespread feeling among nurses in France. In contrast, there are only few reasons why this affects life satisfaction, which is affected, as we have seen, by many other factors.

38Ultimately, all these factors alike highlight the importance of work both as a source of income and as confirmation of status when reporting life satisfaction.

The influence of social relationships

39Nevertheless, beyond work, satisfaction also depends on an individual’s emotional and family life, and friendships—the third dimension defined above. One of the clearest factors under this dimension is linked to the forming of a couple: single people have an average satisfaction score of 5.9, compared with 6.6 for people in a relationship (Pearson correlation of 0.17). In contrast, the presence of children in a household does not have a clear effect on satisfaction, except in single-parent households, where satisfaction falls to 5.4. Other factors also play an important role, such as the number of conversations someone has throughout the day (Pearson correlation of 0.10) and trust in others (Pearson correlation of 0.14). A rich social life is therefore an important factor.

40However, the first two factors considered, linked to income and position in the social hierarchy, weigh more heavily on satisfaction than the third aspect, at least when each is considered in isolation. Essentially, in France, each person is well aware of their status and position, and is sensitive to it. Satisfaction depends on income (as purchasing power and the power to act, or as a source of deprivation when confronted with the products and services on offer), but also on status, going beyond the mere issue of salary (for example, in relation to questions of belonging to a particular class, workplace hierarchy, feelings of upward social mobility, of difficult working conditions, job insecurity, etc.) by positioning each person on one side or the other of inequality and by feeding feelings of being disadvantaged in various ways. Nevertheless, all these factors also help to forge each person’s judgment about social justice in the country.

41It is therefore logical to think that there is a relationship between subjective well-being and feelings of social justice. We have already seen that there seems to be a strong relationship between what we call microjustice—feeling that you are personally treated fairly or unfairly (considered in the survey in terms of salary)—and subjective well-being. But we now turn to feelings of macrojustice, that is, judgments about justice in society as a whole. However, before examining in detail how this could be related to life satisfaction, we consider how it is expressed in the PISJ survey.

Feelings of social justice

42At the end of the questionnaire, the respondents were asked to state whether they think that French society is mostly just or unjust. Here, as in a number of other surveys, a strong majority (60%) said it is mostly unjust. Women judged the situation more negatively than men, and middle-aged people (particularly those between 50 and 60 years) were more critical than younger or older people. However, these differences between the sexes and age groups are not substantial. Socioeconomic status, in contrast, has a very clear impact. The higher someone’s income or the higher their level of education (the two being linked), the more likely they are to think that French society is just (Amadieu and Demeulenaere 2011). Socioprofessional category also plays an important role. Self-employed people judge French society to be more just than employed people, but the higher up the hierarchy of employed professions a person is, the more likely they are to consider French society just. As such, executives consider French society to be mostly just (55%), compared with workers (particularly unskilled workers) or employees, who find it unjust (65%). Here there is not only a scale of feelings of injustice, but a complete inversion of the majority. The same applies if we consider the situation in relation to employment: unemployed people have feelings of injustice roughly equivalent to those of workers. Nationality also has a considerable effect. French people whose parents were both born in France are the most critical, whereas those who have one parent who was born abroad are less critical. The majority of foreign nationals, in contrast to these two categories, find French society mostly just. Here it is plausible that the latter, having more recently arrived in France, compare the situation in France to that in their countries of origin, whereas the judgments of French people may be more idiosyncratic. The respondents’ objective position in the professional hierarchy (or in relation to employment) is, moreover, correlated with their subjective assessment of their position in this professional hierarchy. The more they feel that they belong to an affluent class, the more likely they are to think that French society is just. Here, once again, we observe an inversion of the majority. Those who say that they belong to the upper or upper-middle classes find French society just, whereas the majority of those who think that they belong to the lower-middle classes or, even more so, the working or marginalized classes, find it unjust. It should be noted that this inversion of the majority happens within the middle class, between those who position themselves relatively higher and those who position themselves relatively lower. The former judge French society to be just (60%), whereas, conversely, the latter find it unjust (60%). As such, there are two contrasting perspectives and this contrast becomes particularly strong and striking between the upper classes who consider French society to be just (68%) and the working classes who describe it as unjust (73%).

43Political beliefs also play a role given that only those who state that they are moderately right-wing consider the situation to be mostly just. Everyone else, from the extreme left to the extreme right, including the moderate left and the center, are more likely to think that the situation is generally unjust. However, a person’s political beliefs do not alter the contrast noted above between the upper classes and the working classes, whether considered in “objective” or “subjective” terms. Within the moderate right, for example, the majority of the working classes consider French society to be unjust. Moreover, although there is not strictly an inversion of the majority, the difference between the judgments of the upper classes and working classes remains significant (in the order of 15–20%). Without dismissing its impact, it is clear that political ideology only alters the influence of socioeconomic status on feelings of macro social justice to a limited extent, or not at all.

44These feelings are not unrelated to the assessment of justice or injustice as regards personal remuneration. Unsurprisingly, those who consider themselves unfairly paid are more likely to think that French society is unjust, and vice versa. But again, the impact of professional status is not dulled by the correlation between micro and macrojustice. Whatever they think of their remuneration, executives retain a propensity to judge society as just. And conversely, the working classes remain critical of French society, even if they consider themselves fairly paid.

45Although, on the one hand, feelings of macrojustice are related to feelings of microjustice, on the other, the continued influence of social class demonstrates that the explanation for the responses to this question must go beyond simply framing this in reference to the defense of personal interests (in terms of an individual’s judgments about their own remuneration) or likewise simply as an issue of adherence to a particular political ideology. It is rather a general feeling that is expressed, one not determined simply on the grounds of utilitarian or ideological considerations. It is general, in that it is certainly comprised in part of these dimensions, but it is not reduced to them. In turn, it is not merely membership of an income, educational, or professional category (all these indicators of socioeconomic status being correlated) that alone “mechanically” determines feelings of justice. Even if it is necessary to speak in relative terms, as we have done, about the roles of politics or personal interests, they are not inconsequential. Accordingly, although we can reject automatic explanations or those that rely on an abstract causality that makes the subject simply the result of a socioeconomic structure or ideological superstructure imposed on them, without them having any say in it, it is still relevant to ask how we can consider the influence of social class more specifically.

46There are of course several possible approaches. Here we have focused on one in particular, as it brings into play an aspect of the territorial dimension that has been somewhat overlooked until now. It should be noted that inequality is not evenly distributed throughout the country. In his recent work, Christophe Guilluy (2010) proposes, for example, that three large circles be used to broadly divide the territory. Firstly, there are major cities and their city centers, which are connected to the globalization movement and benefit from it. These urban centers were traditionally occupied by heterogenous social classes, both bourgeois and workers. But increases in house prices have led to gentrification, pushing the working classes to the periphery and replacing them with the upper-middle classes. The second circle is composed of suburbs (banlieues) occupied by the lower or lower-middle classes, and is home to a large share of immigrants. Lastly, the third circle, rural and peri-urban, has continued to spread geographically for physical reasons, as it consists mainly of residential areas. This circle has had to take in both those fleeing the insecurity of Zones Urbaines Sensibles (ZUS) [Sensitive Urban Zones] and those who have been economically displaced from city centers. Nevertheless, the working classes (workers and employees), around which this zone developed demographically (the rural areas no longer being dominated by farmers), have had to suffer the consequences of deindustrialization with the associated direct or indirect job losses. The France of “plans sociaux” (or “job protection programs” which often involve redundancies) is more evident there than anywhere else. In addition to these job losses or risks thereof, public services are being restructured. This has led, in many cases, to growing remoteness (post offices, maternity units, etc.). While politics in the city is focused on the suburbs and enables the genuine renovation of buildings, the peri-urban areas do not benefit from anything similar. Prestigious educational establishments remain the preserve of city centers. Suburban areas have benefitted from Zones d’Éducation Prioritaire (ZEP) [Priority Education Zones] and their increased funding, while peri-urban and rural areas have had access to neither. Over time, qualitative interviews have uncovered feelings of abandonment among those in the rural or peri-urban working classes who feel increasingly left behind by the state and who no longer see themselves in the discourse of the political organizations that have, for the past thirty years, been taking turns at the reins of the state.

47In this context, summarized in very broad terms, can we identify in the PISJ survey a cross effect of area of residence and socioprofessional category that would substantiate this hypothesis that strong feelings of abandonment, and ultimately of injustice, are particularly widespread among the rural and peri-urban working classes? The question of macrojustice that we have studied here is a relevant indicator in this regard. It is above all in rural communities that people are more likely to think that society is unjust. If we also consider socioprofessional category, we find that this negative judgment about society as a whole is most visible among the working classes (workers and employees, and even more so among unskilled workers) living in these rural communities or in towns or villages (in contrast to senior executives in these areas who continue to find the situation more just). These feelings of injustice are conversely much less prominent among workers and employees in suburbs or major cities. It seems, therefore, that French society is perceived as mostly unjust above all by these working-class people who, living in the countryside or peri-urban areas comprised of towns and villages, suffer the full force of the effects of deindustrialization, the growing remoteness of public services, and the increasing burden of the changes in the price of energy required for transport and heating.

48This is incidentally a view specific to the domain of macrojustice. The category of residential area does not affect how people judge the fairness of their own remuneration, and above all there is no such cross effect of residential area or of social class with regard to how the respondents judge inequalities. Here these two variables have neither a significant effect in and of themselves, nor a cross effect. For example, it is not the working classes or rural or peri-urban populations who are most likely to think that income inequality is high in France. This finding may seem surprising, in so far as it is true that we often tend to assimilate, without consideration, inequality and injustice, whether in terms of “objective” measures or perceptions. The respondents here remind us that this assimilation cannot be taken for granted. Judgments about social justice are the product of a combination of factors that go far beyond merely observing the state of inequalities, economic or otherwise. Feelings of abandonment or social and cultural exclusion could also be relevant (strictly, we can only hypothesize here), which could even, if perpetuated, manifest themselves politically in an extreme vote.

Four examples of cross-tabulation between subjective well-being and feelings of social justice

49In this context, it is unsurprising that we find a relationship between feelings of macrojustice and those of well-being (Pearson correlation of 0.22; cf. Table 3). Indeed, as we have seen, life satisfaction articulates a general judgment that takes into account material, social, and emotional conditions and in which the assessment of one’s own situation relative to that of others is important. Nevertheless, this assessment clearly has an influence on feelings about justice in society as a whole. Moreover, Rawls links the assessment of social justice to the lot of the least well off: ultimately, in a just society (called “well-ordered” by Rawls), the least well off should not experience any deprivation or as little deprivation as possible, not because they adapt to their lot, but because their lot is justified in their eyes—it is the equitable result of social cooperation. In fact, dissatisfied people have both a lack of resources and deprivation: in “Rawlsian goods,” they can only surmise that society is unjust. Conversely, satisfied people instead find that society recognizes their value and that such a society must, presumably, be just. The consonance largely observed between well-being and feelings of justice is therefore to be expected.

50Yet, it is interesting to consider dissonant cases, that is, those cases in which personal satisfaction and feelings of social justice seem to be at odds. It thus appears that there are two further possibilities. One in which dissatisfaction does not prevent someone from deeming society to be just, as predicted, for example, by the just-world hypothesis (Lerner 1982; Bénabou and Tirole 2006). And another, conversely, in which the feeling that society is unjust does not prevent someone from thinking that they have done fairly well out of it and expressing feelings of satisfaction (Dubet et al. 2006). In all, we will therefore examine the following four cases: satisfied people who consider society just; dissatisfied people who consider society unjust; satisfied people who consider society unjust; and, finally, dissatisfied people who consider society just. In order to construct this typology, we have considered dissatisfied people as those with a satisfaction score less than or equal to 6 (which is the mode of the distribution) and satisfied people as those with a score higher than 6.

51Empirically, the four cases do occur, but not at all to the same level. A fairly clear relative majority of 33% lean more toward both the injustice of society and personal dissatisfaction. This is followed by the 27% who find society just while also considering themselves satisfied. Then come the 25% who are satisfied while judging society to be unjust. Finally, we have the smallest number, 15%, who, although dissatisfied, find society to be fairly just. As is to be expected given the strong correlation between the two variables, the two most frequent cases are the consonant cases. Nevertheless, satisfied people who state that society is unjust still represent a quarter of the sample. It seems therefore that the positive correlation between satisfaction and feelings about macrojustice relies to a lesser extent on the specific contribution of satisfied people. This correlation relies above all on the clear tendency of dissatisfied people to judge society as unjust.

52Different social or demographic categories correspond to these four combinations (cf. Table 4). Dissatisfied people who judge society as just are distinguished by the fact that often both of their parents were born abroad and the majority of them live in major cities, whereas satisfied people who find society unjust often have parents who were both born in France and may live in rural areas. Moreover, they also have moderately higher incomes and feel that they belong to the upper-middle classes. Yet most striking here is the contrast between the judgments of people linked to a recent immigration and those for whom this is not the case (insofar as this contrast is not found when the correlation between the two variables is positive). The former, although in a less advantaged position, maintain a certain trust in the justice of society, perhaps because their reference point is also, or still, based in their society of origin, whereas the latter are much more critical of society, all the workings of which they know better, even if they say that they have done fairly well out of it (this fits with the findings of François Dubet et al. in Pourquoi moi?, 2013).

Table 4

Typology of cross-tabulation between life satisfaction and feelings about justice in French society by different explanatory variables^*

Dissatisfied and Just Dissatisfied and Unjust Satisfied and Just Satisfied and Unjust Sex Male Female Male n. s.* Age n. s. n. s. n. s. n. s. Socioprofessional category n. s. Worker, employee Senior executives, intermediate professions, self-employed n. s. Education n. s. Primary to secondary University n. s. Employment n. s. Unemployed Employed Employed Standard of living (on average) €1,300 €1,200 €1,800 €1,650 Subjective social class n. s. Disadvantaged, working class Upper, upper-middle class Upper-middle class Parents’ place of birth Both parents born abroad n. s. n. s. Both parents born in France Area of residence Major city n. s. n. s. Rural Marital status n. s. Divorced, separated Married n. s. Politics n. s. Left, center Right n. s. Headcount n = 246 n = 554 n = 446 n = 421 Percentages 14.8% 33.2% 26.8% 25.3%

Typology of cross-tabulation between life satisfaction and feelings about justice in French society by different explanatory variables^*

* = not significant; the categories not presented in this table are without significant effect.

53As regards the consonant cases, we find a clear contrast between the disadvantaged categories (low income, low educational level, often unemployed, workers or employees, those with feelings of marginalization) and the more privileged categories (high income, high level of education, employed, executives, intermediate or liberal professionals, self-employed people, those who feel they belong to the upper classes or at the very least the upper-middle classes). The former (the majority) express strong feelings of dissatisfaction and injustice; whereas among the latter, conversely, satisfaction is coupled with feelings of social macrojustice. It should be noted that while the former are slightly more likely to be women, the latter are more often men. Beyond that, marital status also plays an important role. Divorced or separated people make the same judgment as the disadvantaged categories, while the judgment of married people is the same as that of the privileged categories. The breakdown of a relationship leads therefore not only to dissatisfaction, but also to a more critical judgment of social justice, whereas being in a relationship leads to subjective well-being, which is coupled with a positive judgment about justice in society as a whole. Finally, it is apparent that politically left-wing or centrist people are more dissatisfied and lean more towards considering society unjust, whereas those who state that they are right-wing are more likely to be more satisfied with their life, while considering society more just.

54In short, we therefore have a very pronounced divide between, on the one hand, women, divorced people, the disadvantaged categories, and left-wing or centrist people who are dissatisfied with their lives and social justice, and on the other hand, men, married people, the privileged categories, and right-wing individuals who state that they are satisfied with their lives and social justice.

55Overall, therefore, life satisfaction and the feeling that society is mostly just go hand in hand. When this is not the case, two types of considerations emerge. On the one hand, we have upper-middle class French people, with French parents, living in rural communities, who are satisfied while remaining critical of society. Well integrated, they seem particularly sensitive to any form of arbitrariness as they fear being downgraded or, especially if they live in rural or peri-urban areas, they experience the feeling of abandonment described above. As such, although they feel they are upwardly mobile, they think—like the dissatisfied group that judges society to be unjust—that poverty and exclusion can affect everyone, and not just certain vulnerable groups. This feeling of fragile satisfaction leads them to more easily put themselves in the place of the least well off and, consequently, to judge society as unjust. Conversely, on the other hand, we above all find immigrants or people whose parents were born abroad who seem more forgiving of French society, even though they are dissatisfied with their lot. Poorly integrated, they are perhaps more inclined to accept the idea of being treated differently to others. As such, when asked to judge, in the PISJ survey, the situation “an unskilled worker works hard for their whole life without ever earning more than minimum wage,” they find this a little less unjust than the other social groups. Conversely, this highly relative acceptance of “double standards” seems tied to the promise of integration in the future. It is presumably for this reason that they are in turn more inclined than the other social groups to say, in response to another question in the survey, that the situation “an immigrant has been working in the construction industry for ten years without being registered, and is going to be sent back to their country of origin” is unjust. We are far from the hypothesis of a “belief in a just world” shared by all.

56So far, we have observed the existence of three strong relationships: between feelings of micro and macrojustice, between well-being and microjustice (fair remuneration), and between well-being and macrojustice (social justice). Nevertheless, this last relationship, which we considered in detail, is susceptible to change depending on an individual’s opinion about the fairness of their own remuneration. A log-linear model demonstrates that although the relationships linking each of the three variables to one another are clearly present, there is not a three-way interaction between these three variables. This means that the relationship between subjective well-being and feelings of macrojustice does not depend on what people think of their own remuneration. This relationship remains equally strong among those who consider themselves unfairly paid as among those who think that their pay is fair. We cannot therefore assert that the correlation between well-being and feelings about justice in society as a whole are merely the result of projecting how people judge their own salary level. This correlation goes further and resists purely utilitarian reasoning. From this perspective, macro social justice appears overall to be a factor that affects personal well-being, that is, it cannot be reduced to its other undeniable components.

57This conclusion is further strengthened by the fact that, although there is a clear relationship between feelings of justice and the perception of inequality, this perception does not however affect life satisfaction. In the PISJ survey, we asked the respondents to judge the level of inequality in France in twelve domains (relating to the economy, people’s identity, or various threats), by assessing this level on a scale from 1 to 10 for each domain. Income inequality appears from this perspective to be one of the highest. And it is fairly clear that the more someone thinks that inequality is high in France, the more likely they are to find French society unjust. However, it is not the same sociodemographic characteristics of the interviewees that explain the responses to these two types of questions. To take just one example, while the socioprofessional category of interviewees has an effect (see above) on their feelings of justice, it does not have any effect on their perception of inequalities. To a certain extent, the difference between inequality and injustice can be observed here, given that feelings of justice are related to subjective well-being, while the perception of the level of inequality (in particular, but not only, as regards income) is not. In essence, whereas the understanding of personal well-being seems to require that feelings of justice be taken into account, this is not the case concerning the perception of inequality.

Conclusion

58The Easterlin Paradox put well-being in the spotlight, suggesting that wealth does not lead to well-being at societal level. Even though this hypothesis has since been shown to be wrong, the subject remains of interest. Beyond wealth and GDP, well-being has been accepted as an important dimension for judging the success of a society. At present, there is an abundance of literature on this subject: journals are dedicated exclusively to it; frequent surveys are conducted to assess life satisfaction in most countries around the world; and an institution like the OECD has a website (www.oecdbetterlifeindex.org) dedicated to building its own indicator of well-being.

59The determinants of life satisfaction are therefore starting to be better understood. At the individual level, good health, sufficient money, good working conditions with career advancement, and a rich and full emotional and social life are necessary. The above analysis of the PISJ survey has reached the same conclusions: that financial security and feelings of professional, emotional, and social success are necessary. Furthermore, as we have seen, comparisons with others play an important role in the assessment of one’s own successes and failures.

60However, to date, little attention has been paid to the relationship between feelings of social justice and life satisfaction. Nevertheless, we have been able to observe that such a relationship seems to exist and that this is not by chance. As suggested by Rawls’ theory of justice, feelings of justice are in part related to the satisfaction of the least well off and, likewise, to the feeling of being treated with respect and in equal dignity. The correlation observed here between satisfaction and opinions about justice is therefore to be expected. As regards all the indicators of well-being used in the specialist literature, it therefore seems appropriate to finally include feelings of social justice too. This conclusion is supported by the fact that in Simon Langlois’ study on Quebec (published in this issue of L’Année Sociologique), an equally strong link is found between feelings of well-being and justice, despite the difference in context, given that in contrast to the French, the majority of Quebeckers consider their society mostly just.

61The fact remains that, for Rawls, well-being is not justice. Of course, both share certain conditions, such as having a decent income or the capacity to act and develop professionally, emotionally, and socially. But, as a social objective, the two are clearly distinct: the well-being approach remains somewhat uncritical of the social distribution of the “goods” that it promotes, and it struggles to define an organizing principal for different individuals’ conceptions of the good. The social justice approach tends instead to make justice the organizing principal and the basis for understanding social cohesion, in which well-being has a place. Consequently, these two approaches can make very different assessments of, for example, a community that cultivates social homogeneity and an open society: the satisfaction of being with similar people and the ability to put oneself in another’s place can lead to differences in opinion between the well-being and social justice perspectives.

62This difference is not neutral if we think, as Rawls’ quotation in the introduction reminds us, that justice necessarily has priority over the good. This does not, of course, mean that researching well-being should be rejected from a moral standpoint, but that it should be done under the auspices of justice, which here is always also an end and never merely a means.